What it does
Firecrawl MCP (by Firecrawl / Mendable) turns any website into clean, LLM-ready markdown — single-page scrape, whole-site crawl, URL mapping, web search, and JS-rendered interaction — through one server, for agents needing reliable data from messy, dynamic sites. Scrape/map/crawl/search are deterministic API calls; extract and agent tools are LLM-backed.
Example prompts
- Scrape this product page and give me the title, price, and description as JSON.
- Crawl this documentation site and return clean markdown for every page.
Tools provided
- firecrawl_scrape — Scrape one page to clean markdown.
- firecrawl_map — Map a site URLs.
- firecrawl_crawl — Crawl a whole site.
- firecrawl_search — Search the web.
- firecrawl_extract — Extract structured data (LLM-backed).
What you'll need to connect
This agent will ask you for the following. You enter them when you connect — they're encrypted and never shared with the creator.
- Firecrawl API Key · optionalRequired for stdio and the keyed hosted path; the /v2/mcp path has a keyless free tier.Get this from mcp.firecrawl.dev's account or API settings.Paste the value as a single line.Only sent to: mcp.firecrawl.dev, api.firecrawl.dev
How you're protected
FindAgent runs these safety checks on every agent automatically. They're always on and can't be turned off.
- Prompt-injection scanning
Every request is checked for known prompt-injection and jailbreak attempts before the agent runs. This is always on.
- Secret-leak scanning
Every response is scanned for leaked API keys, tokens, and other secrets before it reaches you. This is always on.
