connect/firecrawl
Bounded Firecrawl search, extraction, mapping and crawl tools
Tools
Readme
Firecrawl
Scrape a page, search the web, discover URLs, or start a small crawl or batch scrape using the deployment’s Firecrawl account.
Call
With Crab signed in to Cloud, replace owner/firecrawl with the address of a running deployment you can access. If this page only offers a template, deploy a copy first.
crab call owner/firecrawl scrape '{"url":"https://example.com"}'
Returns the page data. Use search for up to ten results, map for URL discovery, and crawl or batch_scrape for background jobs.
On a callable deployment, you can also open a tool on this page to call it, or connect your agent through CrabTalk MCP. Calls use this deployment’s provider credentials.
Deploy your own
If this page offers Deploy your own, use it to create a private copy with your own credentials and state. The CLI equivalent is:
crab deploy --from owner/firecrawl
Set FIRECRAWL_API_KEY through the copy’s credential setup. Scraped pages and search results consume the connected Firecrawl account’s credits.
Template deployment needs a published image. To build a modified version or deploy when no template is published, use the source instructions.
Usage notes
Crawl and batch submission return a job ID. Call crawl_status or batch_scrape_status explicitly and pass skip from the returned next URL for another page. Results retain status, page data and creditsUsed.
Search, crawl and batch inputs allow at most ten results, pages or URLs. Map defaults to 1,000 links and accepts a smaller limit. Domain/path filters allow at most 20 entries.
Scrape and batch support Markdown, HTML and links. PDF parsing and charged extraction add-ons are disabled. Agents, monitors and account operations are not included.
Calls do not automatically poll, paginate or retry. A success: false response is returned as an error.