Firecrawl

Crawl and scrape publicly accessible web pages and extract structured data from them.

Firecrawl (opens in new tab)See the Firecrawl connector page

Crawl and scrape publicly accessible web pages and extract structured data from them.

View config snippet
{
  "mcpServers": {
    "firecrawl": {
      "command": "npx",
      "args": ["-y", "firecrawl-mcp"],
      "env": {
        "FIRECRAWL_API_KEY": "your_firecrawl_api_key"
      }
    }
  }
}

Authentication: API key

Capabilities

  • Crawl a URL, with filtering and extraction options.
  • Scrape a single publicly accessible URL.
  • Extract structured data from a page.
  • Check or cancel a crawl job’s status.

Limitations

  • Self-hosted, not a hosted Anthropic-vetted connector — none of the directory-level trust or OAuth guarantees apply.
  • Unknown whether the hosted and self-hosted variants expose identical tool sets.

3 skills and agents use this connector