MCP MCP server · io.github.firecrawl
io.github.firecrawl/firecrawl-mcp-server
MCP server for Firecrawl — web search, scraping, and biomedical/arXiv paper search.
- host stars
- 7,472
- 30-day movement
- +269/day
- Related entries
- 62
- Connections
- 19
npmlocalstdio
An MCP server exposing Firecrawl's web search, scraping, and mapping capabilities to AI agents, plus a set of biomedical/arXiv paper research tools and monitor management tools. It runs over stdio by default, with a hosted keyless option and a search-only endpoint.
Use it when you want an MCP-compatible agent to fetch live web content, search a coding-agent-oriented index, or query arXiv papers without building custom integrations.
Use it to
- Scrape a known URL into markdown or schema-matched JSON
- Search the web and retrieve full page content
- Search, inspect, and read arXiv/biomedical papers
- Map a site's URLs before scraping
- Run and check Firecrawl monitors
For Developers wiring Firecrawl into MCP-compatible AI agents
- Runs
- installed — on your machine
- Transport
- stdio
- Tools
- 19
- Packaged as
- npm:firecrawl-mcp
- Install
- npx -y firecrawl-mcp
- Needs
- 1 environment variable, including secrets
- Version
- 3.24.0
- Last release
- 2026-08-12
topicsmcpweb-scrapingsearcharxivfirecrawlagent-tools