MCP · Project

Firecrawl MCP

Firecrawl MCP exposes web search, page scraping, site crawling and document parsing to MCP clients.

+ Follow

Builder facts

  • Keyless scope. The hosted Streamable HTTP endpoint offers rate-limited search, scrape and parse without an account credential. Source ↗
  • Cloud access. OAuth and API-key sessions can expose additional tools, subject to plan and team policy. The credit usage tool reports remaining team credits and historical use. Source ↗
  • Local runtime. The open-source server uses Node.js 22 or newer and supports stdio and Streamable HTTP. It can use the Firecrawl cloud API or a self-hosted API. Source ↗
  • File parsing. Direct local-file parsing requires a self-hosted API. The hosted service uses a signed upload handoff. Source ↗

Sources checked

Publisher update ·

Firecrawl MCP v3 service update ↗

Firecrawl announced automatic fallback across data providers in its v3 MCP service.

Change timeline

0 stories

No verified releases in this timeline yet.