Firecrawl MCP Server logo

Firecrawl MCP Server

by firecrawl

Searches, scrapes, crawls, maps, parses, and monitors web content through Firecrawl over MCP.

Install Firecrawl MCP Server

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "firecrawl-firecrawl-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "firecrawl-mcp"
      ],
      "env": {
        "FIRECRAWL_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

Install not yet sandbox-tested— the server itself is confirmed live (we called its tools/list endpoint). The stdio install command is a separate, ongoing check.

Overview

firecrawl/firecrawl-mcp-server connects MCP clients to Firecrawl for web search, page extraction, crawling, document parsing, browser interaction, and structured research. It supports known-URL scraping, URL discovery, multi-page crawls, asynchronous agent jobs, recurring change monitors, and scientific-paper searches. Use the hosted MCP endpoints for keyless or OAuth access, or run the package locally over stdio or Streamable HTTP with a Firecrawl API key. Reach for it when an agent needs current web evidence, content from JavaScript-rendered pages, structured fields, or controlled collection across multiple pages.

Use cases

  • Search the web for current information and retrieve source pages
  • Extract defined fields from known URLs as structured JSON
  • Crawl bounded sections of a documentation site
  • Parse PDFs and office documents into agent-readable content
  • Monitor pages for changes and review historical diffs

Key features

  • Web, news, image, developer, and research search
  • Markdown, HTML, links, screenshots, and schema-based extraction
  • Site mapping and bounded multi-page crawling
  • PDF and document parsing
  • Live browser interaction and asynchronous research jobs
  • Recurring monitors with change tracking and notifications

Tools25

What Firecrawl MCP Server lets your agent do. Expand a tool for its parameters.

Verified live~8.1k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

53/53 checks healthy over the last 91d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 4d ago via OSV.dev · firecrawl-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars7.8kStargazers on the source repository.
npm downloads616kPackage downloads in the last 30 days.
Last commitTodayMost recent push to the default branch.
Tools exposed25Callable tools this server registers over MCP.
Directory activity3 installsConfig copies, upvotes, and views on AllMCPs.

How Firecrawl MCP Server works

What firecrawl/firecrawl-mcp-server does

firecrawl/firecrawl-mcp-server exposes Firecrawl web and data-extraction operations through the Model Context Protocol. It can search web, news, and image sources; retrieve a known URL as Markdown, HTML, links, screenshots, branding data, targeted answers, or schema-shaped JSON; and process content from JavaScript-rendered pages, PDFs, and other documents.

The server also covers URL discovery and collection at site scale. firecrawl_map lists indexed URLs without retrieving their bodies, while firecrawl_crawl collects content from multiple pages with path, depth, page-count, subdomain, and external-link controls. Research tools search papers, inspect metadata, find related works, and retrieve relevant passages. Developer search covers indexed GitHub issues, pull requests, READMEs, and selected documentation.

How it works

For a known page, call firecrawl_scrape. Begin with firecrawl_search when the task starts with a query, or use firecrawl_map when the target site is known but the relevant URL is not. Search results can include fetched page content through scrapeOptions, although fetching every result bills each page as a scrape. Use firecrawl_parse for local or uploaded HTML, PDF, Word, RTF, OpenDocument, and spreadsheet files.

Long-running work is exposed as jobs. firecrawl_agent returns a job ID for asynchronous, multi-source research; poll firecrawl_agent_status until completion or failure. Crawls can be checked with firecrawl_check_crawl_status. Browser interaction uses a live session that can be continued with a scrapeId and closed with firecrawl_interact_stop.

Setup and configuration

The hosted MCP endpoint is https://mcp.firecrawl.dev/v2/mcp. Its keyless tier supports scrape, search, and parse with rate limits; other operations require a key. OAuth connections use https://mcp.firecrawl.dev/v2/mcp-oauth, while API-key clients use the standard endpoint with an Authorization: Bearer <FIRECRAWL_API_KEY> secret. A separate search-only endpoint is available at https://mcp.firecrawl.dev/v2/mcp-search.

For local stdio operation, run:

bash
env FIRECRAWL_API_KEY=fc-YOUR_API_KEY npx -y firecrawl-mcp

The package can also be installed globally with npm install -g firecrawl-mcp. Set HTTP_STREAMABLE_SERVER=true to use local Streamable HTTP instead; the local URL is http://localhost:3000/mcp. Cursor and Windsurf configurations use npx -y firecrawl-mcp and pass the API key through the environment.

Tools and capabilities

  • Search web, news, image, developer, GitHub, government, and research sources.
  • Scrape pages into readable content or schema-conforming JSON.
  • Map site URLs and crawl bounded groups of pages.
  • Parse supported documents, including PDFs with page limits.
  • Click, type, fill forms, navigate, or run bounded browser code.
  • Launch structured research jobs and inspect scientific papers.
  • Create, update, run, inspect, and delete recurring content monitors with change results and notifications.

Limitations and notes

Firecrawl may reuse recently indexed content. Set maxAge: 0 when a live fetch is required; a successful response alone does not prove that the represented page state is current. Search-attached page retrieval uses its own reuse behavior.

Interactive browser actions can create persistent effects on a live site, including form submissions. Each interaction call performs one prompt or code turn; sessions can persist through scrapeId, but they are not a client-driven, step-by-step browser loop.

Crawls can produce large responses, so limits should be set when full-site coverage is unnecessary. Safe mode does not support webhooks or interactive scrape actions. The hosted keyless endpoint exposes only a reduced tool set, and the search-only endpoint has a fixed scope.

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Firecrawl MCP Server

firecrawl/firecrawl-mcp-server is an MCP server that connects MCP clients to Firecrawl for web search, scraping, crawling, mapping, document parsing, browser interaction, research, and monitoring. Its main tools include firecrawl_search, firecrawl_scrape, firecrawl_crawl, firecrawl_map, firecrawl_parse, firecrawl_agent, and monitor operations.

Install it with `npx -y firecrawl-mcp` and provide `FIRECRAWL_API_KEY` for local authenticated use. You can also install it globally with `npm install -g firecrawl-mcp`, connect to the hosted MCP endpoint, or configure local Streamable HTTP with `HTTP_STREAMABLE_SERVER=true`.

Yes, it works with Cursor through an MCP server configuration using `npx -y firecrawl-mcp` and a `FIRECRAWL_API_KEY` environment value. The setup instructions target Cursor MCP settings and require Cursor 0.45.6 or newer.

Call `firecrawl_scrape` with the URL and choose the desired output format, such as Markdown, HTML, links, screenshots, or JSON matching a supplied schema. Use `maxAge: 0` when the request must fetch live content rather than reuse recently indexed data.

It can crawl multiple pages from a starting URL, with controls for paths, depth, page count, subdomains, external links, sitemaps, delays, and scrape options. Set conservative limits for large sites, then use `firecrawl_check_crawl_status` to inspect an existing crawl.

Related MCP servers

Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewFirecrawl MCP Server AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/firecrawl-firecrawl-mcp-server?style=directory)](https://allmcps.com/mcp/firecrawl-firecrawl-mcp-server)
HTML Embed
<a href="https://allmcps.com/mcp/firecrawl-firecrawl-mcp-server"><img src="https://allmcps.com/api/badge/firecrawl-firecrawl-mcp-server?style=directory" alt="Firecrawl MCP Server on AllMCPs" /></a>