crw

概述

fastCRW是一款为AI代理设计的高性能网页抓取和爬取API，用Rust构建，提供与Firecrawl兼容的API接口。性能指标优于Firecrawl，仅消耗6MB内存，速度提升5.5倍。项目支持自托管和云部署两种方式，内置MCP服务器功能，可通过stdio和HTTP传输轻松与AI代理集成。

试试问 AI

装完之后，这里有 3 个你可以让 AI 做的事：

• scrape

  Convert any URL to markdown, HTML, JSON, or links

  试试:Scrape https://example.com for markdown content试试:Get the HTML content from https://example.com试试:Extract structured data from https://example.com
• crawl

  Async BFS website crawler with rate limiting

  试试:Crawl https://docs.example.com with depth 2 and max 50 pages试试:Extract content from all pages of https://example.com试试:Scrape an entire website starting from https://example.com
• map

  Discover all URLs on a site instantly

  试试:Map all URLs on https://example.com试试:Find all pages on https://example.com试试:Get the site structure of https://example.com
• search

  Web search + content scraping with built-in SearXNG

  试试:Search for 'open source web scraper 2026'试试:Find recent news about AI web scraping tools试试:Search the web for content about Rust web scrapers

可对比工具

安装

curl -fsSL https://raw.githubusercontent.com/us/crw/main/install.sh | CRW_BINARY=crw sh
crw setup

pip install crw

brew install us/crw/crw

# 本地模式
claude mcp add crw -- npx crw-mcp

# 云模式
claude mcp add -e CRW_API_URL=https://fastcrw.com/api -e CRW_API_KEY=your-key crw -- npx crw-mcp

npx crw-mcp

crw 对比