# Firecrawl
ID: 01-firecrawl
Page: https://gtmmcps.com/tools/01-firecrawl
JSON: https://gtmmcps.com/data/tools/01-firecrawl.json
Documentation references: https://gtmmcps.com/docs/01-firecrawl/
Category: Data & Enrichment
MCP support: official
Testing: RESEARCHED
API access: free
Last checked: 2026-09-07
Endpoint check: not-probed; date: Not recorded

A web scraping and crawling API that turns any URL or whole site into clean markdown or structured JSON for LLM pipelines, with search, map, crawl, extract, parse and browser-interaction endpoints.

## MCP links
- https://mcp.firecrawl.dev/v2/mcp
- https://docs.firecrawl.dev/mcp-server
- https://github.com/firecrawl/firecrawl-mcp-server
- https://mcp.firecrawl.dev/v2/mcp-search
- https://mcp.firecrawl.dev/v2/mcp-oauth

## Sources
- https://docs.firecrawl.dev/mcp-server
- https://www.firecrawl.dev/pricing
- https://github.com/firecrawl/firecrawl-mcp-server
- https://mcp.firecrawl.dev/v2/mcp

## Research notes
Verified 2026-09-07: POST of an MCP initialize to https://mcp.firecrawl.dev/v2/mcp with no key returned HTTP 200 with a JSON-RPC result (protocolVersion 2025-03-26, serverInfo name beginning "firecrawl-"), which matches the documented keyless mode. The control POST to /zzz-not-a-route on the same host returned 200 with the marketing site's HTML, so the control path is not a discriminator for this host and the JSON-RPC body is the evidence. The repo README lists scrape, map, search, crawl, parse, agent, interact, developer_search, research and monitor tools (14 or more). Firecrawl is a general web-data tool, not a contact database; it is in this category because GTM agents use it as the fetch layer under enrichment, the same reason Exa and Bright Data are here. 2026-09-07: https://mcp.firecrawl.dev/v2/mcp returned 200 with a JSON-RPC initialize result to an MCP initialize POST (https://mcp.firecrawl.dev/v2/mcp).

## Named capabilities
Tool names below are reported by sources. Their presence is not a successful call.
- Arguments: Result; evidence: readme; party: first-party
- Categories: `firecrawl_find_tools`; evidence: readme; party: first-party
- Execute: `firecrawl_scrape`; evidence: readme; party: first-party
- Intent: Tool; evidence: readme; party: first-party
- agent: Multi-source research, unknown or many sites; evidence: readme; party: first-party
- crawl: Multi-page extraction (with limits); evidence: readme; party: first-party
- developer: Programming questions over developer sources; evidence: readme; party: first-party
- find_tools: Alexandria catalogue browsing and URL lookup; evidence: readme; party: first-party
- firecrawl_parse: Parse one supported document into markdown, HTML, links, summary, targeted answers, or JSON matching a schema. Supported inputs include common HTML, PDF, Word, RTF, OpenDocument, and spreadsheet files; PDF parsing can be bounded with `pdfO; evidence: live-list; party: 
- firecrawl_scrape: Scrape one URL and return its content: markdown by default, or HTML, links, screenshots, branding data, a targeted answer, or JSON matching a supplied schema. Use it when the request identifies a page and needs its content or defined field; evidence: live-list; party: 
- firecrawl_search: Search web, news, or image sources and return ranked results with query-relevant highlights. Each web result is a title, URL, and description; use `firecrawl_scrape` on a result URL when the excerpt is not enough.  Authenticated search als; evidence: live-list; party: 
- interact: Interact with a URL or scraped page; evidence: readme; party: first-party
- map: Discovering URLs on a site; evidence: readme; party: first-party
- monitor: Recurring page checks; evidence: readme; party: first-party
- parse: Files and hosted upload refs; evidence: readme; party: first-party
- research: Paper and GitHub repository research; evidence: readme; party: first-party
- scrape: Single page content; evidence: readme; party: first-party
- search: Web search for info; evidence: readme; party: first-party

## CLI evidence
Status: official
- Reported install command: npm install -g firecrawl-cli; source: https://docs.firecrawl.dev/sdks/cli
- Reported install command: npx -y firecrawl-cli@latest; source: https://docs.firecrawl.dev/sdks/cli

Read current vendor documentation before configuring or running commands.