Firecrawl
Service domainWEB SCRAPING
Arcade OptimizedBYOCPro
Arcade.dev LLM tools for reading the web via Firecrawl
Author:Arcade
Version:
5.0.1Auth:No authentication required
12tools
12require secrets
Firecrawl is a web scraping and search API. This toolkit lets agents read web pages, crawl and map sites, extract structured records, and search the open web, developer documentation, GitHub issues, and research papers.
Capabilities
- Page reading: read one page whose URL is already known and return its content.
- Crawling and site mapping: crawl many pages of a site, list a site's URLs without reading them, and check progress, read collected pages, or cancel a running crawl by job ID.
- Structured extraction: describe the fields you need (such as products, prices, or contacts) in plain language and get structured records back. Long extractions keep running and return a job ID for collecting the results later.
- Web search: search the open web and optionally read each result page in the same call.
- Specialized search: search indexed developer documentation and repositories, GitHub issues, or a corpus of research papers drawn from PubMed, bioRxiv, medRxiv, and arXiv.
Secrets
FIRECRAWL_API_KEY: your Firecrawl API key, sent with every request. Sign in to the Firecrawl dashboard and create or copy a key there. Usage counts against your Firecrawl plan's credits and rate limits.
Store the secret in Arcade as described in the Arcade secrets docs, or manage it in the Arcade dashboard.
Available tools(12)
12 of 12 tools
Operations
Behavior
| Tool name | Description | Secrets | |
|---|---|---|---|
Stop a crawl that is still running.
Pages the crawl already collected stay readable by its job id. | 1 | ||
Collect structured records from the web from a plain-language description.
Use this when the answer is fields such as products, prices, or contacts,
rather than the page text itself. An extraction that outruns wait_seconds
keeps running, and the result carries its job_id for collecting the results
later. | 1 | ||
Check how far a crawl has progressed, without fetching its pages. | 1 | ||
Collect the results of an extraction that is already running. | 1 | ||
Read one web page whose URL is already known and return its content.
It reads only the page at the given URL. It does not search the web, crawl
other pages of the site, or pull structured fields out of the page text. | 1 | ||
Read the pages a crawl has collected so far. | 1 | ||
List the URLs on a website, without reading the pages. | 1 | ||
Search developer documentation and repositories, quoting the text that matched.
Use this to answer a question about how a library or API works. It searches
indexed repositories, not the open web. | 1 | ||
Search GitHub issues across indexed repositories.
Only issues are searched; documentation, READMEs, and pull requests are not. | 1 | ||
Search published scientific papers and return their abstracts.
This reads a corpus of paper records drawn from PubMed, bioRxiv, medRxiv,
and arXiv. To find ordinary web pages that happen to sit on academic sites,
use a web search narrowed to research instead. | 1 | ||
Search the web and optionally read each result page in the same call.
Use this when no URL is known yet. It searches the open web rather than a
corpus of scientific papers or indexed code documentation. | 1 | ||
Read many pages of one website and return each page's content.
A crawl that outruns wait_seconds keeps running: the result carries its
job_id for checking progress, reading the pages, or cancelling it later. | 1 |
Last updated on