Build with Berrycrawl
Turn websites into clean data, screenshots, site maps, and brand profiles for AI agents.
Berrycrawl is a web-data API built for agents. Send it a URL and ask for the form your workflow needs: readable Markdown, raw HTML, links, a screenshot, structured JSON, a crawl, or a brand profile.
Make your first request
Choose a workflow
Use an SDK
Read the API reference
Base URL
All public data-plane requests use:
https://api.berrycrawl.com/api/v1Authenticate with a bc_ API key in the bearer header. JSON is the default request and response encoding.
Authorization: Bearer bc_your_api_key
Content-Type: application/jsonPick the smallest operation that solves the job
| Need | Start with |
|---|---|
| One page as Markdown, HTML, links, JSON, or a PDF | POST /scrape |
| A durable image of a page | POST /screenshot |
| URLs from one site without fetching every page | POST /map |
| Pages across a site | POST /crawl |
| Search results from the public web | POST /search |
| Structured facts from one or more pages | POST /extract |
| Logos, colors, description, and brand imagery | GET /brand/retrieve |
Berrycrawl starts with the least expensive reliable fetch and escalates when a site requires browser rendering or a residential IP. You request the result; the platform owns the browser and proxy decisions.
Free test domain
Requests for berrycrawl.com are free, so an agent can validate its integration before spending credits. New workspaces receive 100 credits; paid credits are added through pay-as-you-go top-ups.