Firecrawl MCP Server
Official Firecrawl MCP server: web search, page scraping to Markdown, site crawling and in-page actions
- 7,712 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Sends URLs and queries to an external service
- Consumes paid Firecrawl account credits
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add mcp/firecrawl-mcp-serverDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Sends URLs and queries to an external service
- Consumes paid Firecrawl account credits
This is third-party code. Review the repository files before installing.
What it does
The server exposes Firecrawl tools to the agent. It can fetch a known URL as Markdown or schema-shaped JSON, map a site's links, crawl a section with depth limits and search the web with optional content fetching. There is a tool for clicking and navigating before reading, scheduled page change monitoring, research paper search and an agent mode for multi-step research. It works against the Firecrawl cloud API or a self-hosted instance.
Who it is for. For developers and researchers who need an agent to read live websites as clean content.
Good fit when
- You need docs pages or articles as Markdown for the agent
- You need data from several pages of a site
- A page needs a click or input before it shows content
Not a fit when
- You need a long browser session driven step by step from the client
- You cannot send URLs and page content to an external service
Example request
Scrape the docs.example.com/api section into Markdown, at most 20 pagesMCP
- Transport
- stdio, http, sse
- Authentication
- API key
Environment variables
FIRECRAWL_API_KEYrequired, secret- Firecrawl cloud API key, optional for self-hosted
FIRECRAWL_API_URL- URL of a self-hosted Firecrawl instance
FIRECRAWL_OAUTH_TOKENsecret- OAuth access token instead of a key in stdio mode
Limitations
The cloud needs FIRECRAWL_API_KEY or an OAuth login, and the service has paid plans. The keyless hosted endpoint offers only scrape, search and parse with rate limits. Access to Firecrawl cloud from Russia without VPN is unverified; you can self-host instead.
How to disable. Remove the firecrawl block from your client's MCP config.
FAQ
Can I use it without a key?
Yes, via the hosted https://mcp.firecrawl.dev/v2/mcp endpoint, but only three tools are available and rate limits apply.
Does it work with self-hosted Firecrawl?
Yes, set FIRECRAWL_API_URL; the key becomes optional.
Discussion
No questions or reviews yet.
Related
MCP serversThe Firecrawl command-line client with agent skills: web search, scraping, crawling, URL mapping and page interaction
A Rust server that turns pages into Markdown or JSON: search, scrape, site map, crawl and PDF through MCP, a CLI and a Firecrawl-compatible API
Official Apify MCP server: the agent runs ready-made Apify Store scrapers for social media, maps, search and shops
A unified search MCP server: one interface to Tavily, Brave, Kagi, Exa, Linkup, Firecrawl and GitHub search
An open-source crawler that turns websites into clean Markdown for AI agents: install from PyPI, run in Docker and connect to Claude Code over MCP
An MCP server that opens pages in headless Chromium through Playwright and hands the agent the main text as Markdown, even from JavaScript-heavy sites
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Firecrawl MCP Server (this entry) | yes | API key | 7,712 | |
| Crawl4AI | yes | API key | 86,495 | an open-source crawler that turns websites into clean Markdown for AI |
| Bright Data Skills | yes | 268 | the official Bright Data plugin | |
| fastCRW | yes | not required | 1,192 | a Rust server that turns pages into Markdown or JSON |
README in short
The README explains which tool fits which job: scrape, map, crawl, search, interact, monitoring and research tools. It covers the hosted server with OAuth or key, a separate search-only endpoint, running via npx and configs for Cursor, Windsurf, VS Code and Claude Desktop. It lists environment variables for cloud and self-hosted setups. MIT licensed.