Integrate with Firecrawl, a web scraping and crawling API that turns any website into LLM-ready data. Scrape single pages into markdown or HTML, search the web, map all URLs of a site, run asynchronous website crawls, and extract structured data from pages with an LLM prompt or JSON schema.
Changelog
1 version
v1.0.0
Latest
- Initial release. Adds API-key authentication and components for scraping a URL, web search, mapping website URLs, asynchronous website crawls with polling, structured data extraction with an LLM prompt or JSON schema, retrieving crawl and extract job status, and an authenticated Make API Call passthrough.
Actions
8
Crawl Website
Start an asynchronous crawl of a website and scrape every discovered page into markdown. The job reference is emitted immediately on "out"; Firecrawl calls back when the crawl finishes and the scraped pages are emitted on "done".
Extract Data
Extract structured data from one or more URLs using an LLM prompt and/or a JSON schema. By default the component waits for the extraction to finish and returns the extracted data; turn off waiting to only get the job ID and fetch results later with Get Extract Status.
Find Search Results
Search the web, news or images with Firecrawl and return the ranked results. Returns up to 20 results per search.
Get Crawl Status
Get the status of a crawl job and, when it is completed, the scraped pages. Use it with a job ID returned by Crawl Website.
Get Extract Status
Get the status of an extraction job and, when it is completed, the extracted data. Use it with a job ID returned by Extract Data.
Make API Call
Make a custom authorized call to the Firecrawl API. Use it for endpoints or parameters that dedicated components do not cover.
Map Website
List the URLs of a website, discovered from its sitemap and links. Returns up to 5000 URLs, optionally filtered by relevance to a search term.
Scrape URL
Scrape a single URL and return its content as LLM-ready markdown, optionally together with HTML, links, a summary or a screenshot.