Firecrawl MCP

Agent

Firecrawl MCP

Creator:

mcp_depot

About this agent

Firecrawl's official server for turning websites into LLM-ready data. Agents can scrape single pages, crawl whole sites, map URLs, search the web and extract structured JSON. It runs hosted, with the API key sent as a Bearer header or via OAuth, or locally through npx, and a keyless tier covers basic search and scraping. It handles JavaScript-heavy pages and rate limiting for you.

Connect

MCP endpoint: https://mcp.firecrawl.dev/v2/mcp

Add it to any MCP client that supports remote servers (Claude, Cursor, VS Code, Windsurf):

JSON
{
  "mcpServers": {  
    "firecrawl": {  
      "url": "https://mcp.firecrawl.dev/v2/mcp"  
    }  
  }  
}

It can also run locally:

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

Details

  • Authentication: API key (Bearer header), OAuth, or keyless (limited)
  • Maintainer: Firecrawl (official server)
  • Category: Web Scraping
  • Tags: scraping, crawling, web-data, markdown
  • Popularity: 7.2k searches/mo; 7.6k GitHub stars

Connect to this MCP server

Server URL

https://mcp.firecrawl.dev/v2/mcp
Checking the server…
Add to CursorAdd to VS Code

Claude Code

Run in your terminal. Add --scope user to use it in every project.

Claude Desktop: claude_desktop_config.json

Open it from Settings → Developer → Edit Config, then restart Claude. Or skip the file: Settings → Connectors → Add custom connector, and paste the URL.

Comments & Discussion

Scroll to load comments...

Tags

scraping
crawling
web-data
markdown
mcp

Share

Loading recommendations...