ChatGPT Scraper

This Actor gets answers from ChatGPT and returns structured data. It supports custom prompts, country/region selection, and web search functionality.

What this Actor does:

1. Web Reverse Engineering: Uses web scraping techniques to interact with ChatGPT's web interface
2. Simulates AI Interaction: Automatically navigates to ChatGPT and submits your prompt
3. Waits for AI Response: Monitors the conversation until ChatGPT completes its answer
4. Extracts Structured Data: Captures the complete response including text, references, and other elements
5. Returns JSON Output: Formats all extracted data into a structured JSON format for easy use

Input Parameters:

Example Input:

{
    "prompt": "Top hotels in New York",
    "country": "US",
    "web_search": false
}

Output:

The Actor saves results to a dataset containing comprehensive structured data from ChatGPT's response:

Output Structure:

• answer_text: Plain text response from ChatGPT
• answer_text_markdown: Markdown-formatted response
• answer_html: Full HTML of the response
• answer_section_html: HTML of the answer section
• citations: Array of referenced sources with domains, titles, and URLs
• map: Array of location data with hotel details (name, rating, phone, website)
• links_attached: Array of links referenced in the response
• model: The AI model used (e.g., gpt-5-2)
• prompt: Original input prompt
• input: Original input parameters
• country: Country/region used for the request
• timestamp: ISO timestamp of when the response was generated
• url: URL of the ChatGPT conversation
• web_search_query: Array of search queries used if web search was enabled
• web_search_triggered: Boolean indicating if web search was used
• shopping: Array of shopping results (if any)
• search_sources: Array of search sources (if any)

Example Output:

{
  "answer_text": "Here are some top-rated hotels in New York City, USA, spanning luxury landmarks, iconic stays, and widely recommended favorites for visitors: 🌟 Luxury & Upscale Hotels The Plaza – Legendary luxury hotel on Fifth Avenue near Central Park, famous for elegant rooms and classic NYC glamour...",
  "answer_text_markdown": "Here are some **top-rated hotels in New York City, USA**, spanning luxury landmarks, iconic stays, and widely recommended favorites for visitors:\n\n### 🌟 Luxury & Upscale Hotels\n\n* **The Plaza**  – Legendary luxury hotel on Fifth Avenue near Central Park, famous for elegant rooms and classic NYC glamour...",
  "citations": [
    {
      "cited": true,
      "description": "December 23, 2025",
      "domain": "Forbes",
      "title": "Best Hotels In New York City 2025 - Forbes Vetted",
      "url": "https://www.forbes.com/sites/forbes-personal-shopper/article/best-hotels-in-new-york-city/?utm_source=chatgpt.com"
    }
  ],
  "map": [
    {
      "category": "Hotel",
      "name": "The Plaza",
      "phone_number": "+12127593000",
      "rating": 4.5,
      "review_count": 6766,
      "website_url": "https://www.theplazany.com/"
    }
  ],
  "model": "gpt-5-2",
  "prompt": "Top hotels in New York",
  "timestamp": "2026-02-28T07:12:50.140Z",
  "url": "https://chatgpt.com/?q=Top%20hotels%20in%20New%20York",
  "web_search_triggered": true
}

Features:

• Structured Data Extraction: Captures comprehensive data including text, markdown, HTML, citations, and map locations
• Web Search Support: Can enable web search for real-time information
• Country Selection: Allows specifying country/region for location-specific responses
• Error Handling: Gracefully handles web scraping errors and AI response issues
• Web Reverse Engineering: Uses advanced techniques to interact with ChatGPT's web interface

Usage Tips:

• For best results, provide clear and specific prompts
• Enable web search for time-sensitive or current information
• Use country parameter for location-specific responses

If you need data from other AI models or have specific requirements, feel free to let me know and I can develop a custom Actor for your needs!