WaterCrawl MCP

A Model Context Protocol (MCP) server for WaterCrawl, built with FastMCP. This package provides AI systems with web crawling, scraping, and search capabilities through a standardized interface.

Use WaterCrawl MCP directly without installation using npx:

npx @watercrawl/mcp --api-key YOUR_API_KEY

Configure your Codeium or Windsurf with this package without installing it:

{
  "mcpServers": {
    "watercrawl": {
      "command": "npx",
      "args": [
        "@watercrawl/mcp",
        "--api-key",
        "YOUR_API_KEY",
        "--base-url",
        "https://app.watercrawl.dev"
      ]
    }
  }
}

Run WaterCrawl MCP in SSE mode:

npx @watercrawl/mcp sse --port 3000 --endpoint /sse --api-key YOUR_API_KEY

Then configure Claude Desktop to connect to your SSE server.

• -b, --base-url <url>: WaterCrawl API base URL (default: https://app.watercrawl.dev)
• -k, --api-key <key>: Required, your WaterCrawl API key
• -h, --help: Display help information
• -V, --version: Display version information

SSE mode additional options:

• -p, --port <number>: Port for the SSE server (default: 3000)
• -e, --endpoint <path>: SSE endpoint path (default: /sse)

wc-mcp/
├── src/                   # Source code
│   ├── cli/               # Command-line interface
│   ├── config/            # Configuration management
│   ├── mcp/               # MCP implementation
│   ├── services/          # WaterCrawl API services
│   └── tools/             # MCP tools implementation
├── tests/                 # Test suite
├── dist/                  # Compiled JavaScript
├── tsconfig.json          # TypeScript configuration
├── package.json           # npm package configuration
└── README.md              # This file

1. Clone the repository and install dependencies:

git clone https://github.com/watercrawl/watercrawl-mcp
cd watercrawl-mcp
npm install

2. Build the project:

npm run build

3. Link the package for local development:

npm run dev:link

The project includes tests for both SSE and npx modes:

# Run all tests
npm test

# Run only SSE tests
npm run test:sse

# Run only npx tests
npm run test:npx

Tests require a valid WaterCrawl API key to be set in the .env file or passed as an environment variable.

1. Fork the repository
2. Create a feature branch (git checkout -b feature/your-feature)
3. Commit your changes (git commit -m 'Add your feature')
4. Push to the branch (git push origin feature/your-feature)
5. Open a Pull Request

npm install -g @watercrawl/mcp

npm install @watercrawl/mcp

Configure WaterCrawl MCP using environment variables or command-line parameters.

Create a .env file or set environment variables:

WATERCRAWL_BASE_URL=https://app.watercrawl.dev
WATERCRAWL_API_KEY=YOUR_API_KEY
SSE_PORT=3000                  # Optional, for SSE mode
SSE_ENDPOINT=/sse              # Optional, for SSE mode

The WaterCrawl MCP server provides the following tools:

Scrape content from a URL with customizable options.

{
  "url": "https://example.com",
  "pageOptions": {
    "exclude_tags": ["script", "style"],
    "include_tags": ["p", "h1", "h2"],
    "wait_time": 1000,
    "only_main_content": true,
    "include_html": false,
    "include_links": true,
    "timeout": 15000,
    "accept_cookies_selector": ".cookies-accept-button",
    "locale": "en-US",
    "extra_headers": {
      "User-Agent": "Custom User Agent"
    },
    "actions": [
      {"type": "screenshot"},
      {"type": "pdf"}
    ]
  },
  "sync": true,
  "download": true
}

Search the web using WaterCrawl.

{
  "query": "artificial intelligence latest developments",
  "searchOptions": {
    "language": "en",
    "country": "us",
    "time_range": "recent",
    "search_type": "web",
    "depth": "deep"
  },
  "resultLimit": 5,
  "sync": true,
  "download": true
}

Download a sitemap from a crawl request in different formats.

{
  "crawlRequestId": "uuid-of-crawl-request",
  "format": "json" // or "graph" or "markdown"
}

Manage crawl requests: list, get details, stop, or download results.

{
  "action": "list", // or "get", "stop", "download"
  "crawlRequestId": "uuid-of-crawl-request", // for get, stop, and download actions
  "page": 1,
  "pageSize": 10
}

Manage search requests: list, get details, or stop running searches.

{
  "action": "list", // or "get", "stop"
  "searchRequestId": "uuid-of-search-request", // for get and stop actions
  "page": 1,
  "pageSize": 10,
  "download": true
}

Monitor a crawl or search request in real-time, with timeout control.

{
  "type": "crawl", // or "search"
  "requestId": "uuid-of-request",
  "timeout": 30, // in seconds
  "download": true
}

ISC