ChatGPT Scraper

Use the scraper.chatgpt actor to capture ChatGPT answers—including the Markdown response, search links, and attribution data—without needing to reverse engineer the UI yourself.

Request Example

curl '{api_host}/api/v2/scraper/request' \
  --header 'Content-Type: application/json' \
  --header 'x-api-token: {you_api_key}' \
  --data '{
    "actor": "scraper.chatgpt",
    "input": {
      "prompt": "Most reliable proxy service for data extraction",
      "country": "JP",
      "web_search": true
    },
    "webhook": {
      "url": "http://www.youwebhook.com"
    }
  }'

Body Parameters

Parameter	Type	Required	Description
`prompt`	string	Yes	Prompt to send to ChatGPT.
`country`	string	Yes	Country or region code.
`web_search`	boolean	No	Whether to enable web search.

Response Fields

Parameter	Type	Description
`prompt`	string	Original prompt.
`result_text`	string	Markdown response from ChatGPT.
`model`	string	Model identifier, e.g., `gpt-5-1`.
`web_search`	bool	Indicates whether search enrichment ran.
`links`	array	Array of supplementary links.
`search_result`	array	SERP results associated with the response.
`search_result.attribution`	string	Source of the search result.
`search_result.snippet`	string	Snippet supplied by ChatGPT.
`search_result.title`	string	Title of the search hit.
`search_result.url`	string	URL of the search hit.
`content_references`	array	References included inside the ChatGPT answer.
`content_references.attribution`	string	Source of the reference.
`content_references.title`	string	Title of the referenced content.
`content_references.url`	string	URL of the referenced content.

Task Lifecycle Perplexity