Web scraping, crawling, and data extraction for AI agents. 5 built-in tools: scrape (clean markdown from any URL), crawl (follow links across entire sites), map (discover all URLs on a domain), extract (structured JSON via schema), and search (web search with optional page scraping). 833ms average latency, single binary, 6 MB RAM. Self-host free or use managed cloud at fastcrw.com with 500 free credits. Firecrawl-compatible API. Works with Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP-compatible client.
Overview
fastCRW — Web Scraping MCP Server for AI Agents
Give your AI agent reliable web access. fastCRW exposes 5 tools via MCP:
Tools
- crw_scrape — Fetch a single URL as clean markdown, HTML, or screenshot
- crw_crawl — Crawl an entire site, follow links up to configurable depth
- crw_map — Discover all URLs on a domain via sitemap + link traversal
- crw_extract — Extract structured JSON matching a schema you define
- crw_search — Web search with optional full-page scraping
Quick Setup
Install and run with a single command:
npx crw-mcp
Why fastCRW?
- 833ms average scrape latency
- 6 MB RAM footprint
- Single binary — no browser, no Docker, no dependencies
- Firecrawl-compatible API — drop-in replacement
- Self-host free with no limits, or use cloud at fastcrw.com (500 free credits)
Works With
Claude Desktop, Claude Code, Cursor, Windsurf, Cline, and any MCP-compatible client.
Links
Server Config
{
"mcpServers": {
"crw": {
"command": "npx",
"args": [
"crw-mcp"
],
"env": {
"CRW_API_KEY": "<YOUR_API_KEY>"
}
}
}
}