Overview
What is Watercrawl Mcp?
A Model Context Protocol server that provides AI systems with web crawling, scraping, and search capabilities through a standardized interface, built on the WaterCrawl platform.
How to use Watercrawl Mcp?
Use npx @watercrawl/mcp --api-key YOUR_API_KEY to run without installation. Configure via environment variables or command-line options. Integrate with AI assistants like Codeium/Windsurf (JSON configuration) or Claude Desktop (SSE mode).
Key features of Watercrawl Mcp
- Run directly with npx, no installation required
- Integrates with Codeium/Windsurf and Claude Desktop
- Offers six MCP tools for web data operations
- Scrapes web pages with customizable options
- Performs web searches with depth and time range filters
- Monitors crawl and search requests in real time
Use cases of Watercrawl Mcp
- AI agents scraping web content with fine‑grained control
- Performing deep web searches with language/country preferences
- Managing and monitoring multiple crawl or search requests
- Generating sitemaps from crawl results in JSON, graph, or markdown
- Automating data extraction for AI‑powered workflows
FAQ from Watercrawl Mcp
What is the purpose of Watercrawl Mcp?
It provides AI systems with web crawling, scraping, and search capabilities through the Model Context Protocol, using WaterCrawl as the backend service.
What are the runtime requirements?
Node.js is required. The server can run via npx without installation, or be installed globally with npm install -g @watercrawl/mcp.
Where does scraped data reside?
The server acts as a client to the WaterCrawl API; fetched data is returned to the AI assistant and is not persistently stored by the server itself.
What transports and authentication are supported?
By default, it uses stdio transport. SSE mode is available for Claude Desktop. Authentication requires a WaterCrawl API key, supplied via --api-key or the WATERCRAWL_API_KEY environment variable.
