Firecrawl
Arcade.dev LLM tools for reading the web via Firecrawl
4.0.1Firecrawl Toolkit
Firecrawl is a web-scraping and crawling service; this toolkit lets Arcade agents read, search, and extract structured data from the web by calling Firecrawl's API.
Capabilities
- Single-page and multi-page reading: Scrape one known URL or crawl an entire site, collecting each page's content with configurable depth and wait behavior.
- Site mapping: List all URLs on a website without fetching page content, useful for planning targeted scrapes or crawls.
- Structured data extraction: Describe the fields you want (products, prices, contacts) in plain language and retrieve them as structured records instead of raw page text.
- Web and specialized search: Search the open web, developer documentation and repositories, GitHub issues, or published scientific papers (PubMed, bioRxiv, medRxiv, arXiv) — each via a dedicated tool optimized for that corpus.
- Async job management: Long-running crawls and extractions return a
job_id; dedicated status, data-retrieval, and cancel tools let agents poll or stop jobs at any point.
Secrets
FIRECRAWL_API_KEY — A Firecrawl API key that authenticates every request to the Firecrawl service. Obtain it by signing in to the Firecrawl dashboard, navigating to the API Keys section, and creating a new key. The free tier issues keys with rate-limited access; higher-volume usage requires a paid plan. Copy the key immediately after creation — Firecrawl does not display it again.
Configure secrets in Arcade at https://docs.arcade.dev/en/guides/create-tools/tool-basics/create-tool-secrets or via the dashboard at https://api.arcade.dev/dashboard/auth/secrets.
Available tools(12)
| Tool name | Description | Secrets | |
|---|---|---|---|
Stop a crawl that is still running.
Pages the crawl already collected stay readable by its job id. | 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 | ||
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 | ||
Read the pages a crawl has collected so far. | 1 | ||
Check how far a crawl has progressed, without fetching its pages. | 1 | ||
Collect the results of an extraction that is already running. | 1 | ||
List the URLs on a website, without reading the pages. | 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 | ||
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 | ||
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 Search narrowed to research instead. | 1 |