Firecrawl MCP Server
7,536 starsFirecrawl's official MCP server: a full web-acquisition toolbox — firecrawl_search (query-first search), firecrawl_scrape (URL to clean Markdown/JSON), firecrawl_map (URL discovery), firecrawl_crawl (multi-page crawl with limit and path filters), firecrawl_interact (click/type page interactions), monitor tools (scheduled checks with diff alerts) and credit_usage (quota checks). A full profile registers 26 tools; there is also a keyless hosted endpoint (https://mcp.firecrawl.dev/v2/mcp, 3 tools) and a search-only endpoint (8 tools).
- First published：2024-12-06
- data as of：
- 2026-09-30
- Licence：
- MIT
- Transport：
- stdio / http
- Primitives exposed：
- tools
What it can touch
All fetching executes through Firecrawl's cloud: the URLs and queries you crawl go to Firecrawl's servers, and fetched content enters the model context. The interact tool performs clicks and typing on pages; monitor creates scheduled jobs; credit_usage reads your billing quota. The API key draws down paid credits.
Risk
Three layers: arbitrary fetched pages are an injection vector (far riskier than first-party docs); target URLs and queries are disclosed to a third party (Firecrawl) and may touch site ToS; and autonomous crawl/monitor calls burn paid credits — the existence of the credit_usage tool tells you quota is being consumed.
Install prompt
本地方式：從官方倉庫 README 取得設定片段（以 npx 啟動 firecrawl-mcp，用 FIRECRAWL_API_KEY 環境變數傳入金鑰），加進客戶端設定後重啟。或直接使用 README 記載的託管端點（免金鑰端點只有 3 個工具）。Paste this to your agent. It deliberately contains no tool subcommands — check the tool's official docs for commands.