chatgpt-answer
What ChatGPT answers a logged-out visitor for one question, as one row: the answer as Markdown and as plain text, the sources it cited and the pages its web search read. For brand monitoring and AI-search visibility: does ChatGPT name you, and which pages does it cite. prompt is the question; web_search lets ChatGPT search the web first (default on); country asks from that country. About 30 s to 2 minutes.
Recipes are not open yet.
What this recipe returned on its last run. No key needed to look.
Fields
7 columns on every row. null means the site does not state it.| field | type | meaning |
|---|---|---|
| prompt | The question as it was sent to ChatGPT. | |
| answer | ChatGPT's answer as Markdown. | |
| answer_text | The same answer as plain text. | |
| citations | The sources the answer cites, in order, each {url, title, description}; ChatGPT's tracking parameter is removed from the URL. | |
| sources | Every page ChatGPT's web search read for the answer, each {url, title}; it holds the citations and the pages it read but did not cite. | |
| web_search_used | true when ChatGPT searched the web for this answer. | |
| asked_at | When the question was sent, ISO 8601 UTC. |
null means the site does not state it.
How it behaves
Cost. 1 record. A failed run costs 0.
Freshness 1 h. The same input within 1 h is served from cache at 0 units.
How it answers. Answers at once for one input; a large run goes as a job you poll.
Questions about this recipeQuestions
Short answers. The API reference has the long ones.
Call POST /recipes/chatgpt-answer with your key and prompt and web_search and country. You get rows with the 7 fields above. The recipe is tested against the live site every 6 hours.
1 record. A failed run costs 0. The same input within 1 h costs 0.
Yes. Sign up with Google, GitHub or email for a free key with 1,000 credits every month. No credit card.
The health check runs every 6 hours against the live site. A failing recipe shows as failing on this page until we fix it.
Extra fields, higher volume, or a private copy with a freshness SLA. We build it on Enterprise.
Request a changeMore for research and knowledge
All 5 recipes →Papers on arXiv that match a query, from the arXiv API: title, authors, the first 500 characters of the abstract, the paper's page, the…
One Wikipedia article: its title, short description, summary paragraph, page link, thumbnail, the list of top-level sections, when it was…
Wikipedia articles that match a query, from the public search API: title, article link, a plain-text snippet around the match, word count…
For developers
Get a free API key and run your first crawl.
1,000 credits every month. No credit card. Your self-hosted setup keeps working.
For teams
Talk to us about volume, regions and an SLA.
A 30-minute call with the team that writes the crawler. DPA on request, invoice billing, your choice of US, EU or Singapore.