MCP WebcrawlerAPI integration
MCP (Model Context Protocol) is a protocol that allows AI assistants to connect to external services and tools. The WebCrawlerAPI MCP server enables AI assistants like Claude Code to scrape web content directly during conversations.
Installation
Using npm
npm install -g webcrawler-mcp
Using npx (no installation required)
npx webcrawler-mcp
Setup
1. Get your API key
Get your API key from the WebCrawlerAPI dashboard (opens in a new tab).
2. Set environment variable
export WEBCRAWLER_API_KEY="your-api-key-here"
3. Run the server
The MCP server can run in two modes:
Standard mode (stdio):
npx webcrawler-mcp
HTTP mode:
export USE_HTTP=true
export PORT=8080
npx webcrawler-mcp
Configuration
The MCP server supports the following environment variables:
WEBCRAWLER_API_KEY
(required): Your WebCrawlerAPI access keyUSE_HTTP
(optional): Enable HTTP transport modePORT
(optional): HTTP server port (default: 8080)
Using with Claude Code
Once the MCP server is running, Claude Code can automatically discover and use the webcrawler-scrape
tool to scrape web content.
The tool accepts these parameters:
url
(required): The webpage URL to scrapeprompt
(optional): Specific information to extract from the page
Example usage in Claude Code:
Please use webcrawler-scrape to scrape https://example.com and extract the main article content
Claude Code will automatically use the MCP server to scrape the webpage and return the content as markdown.
Features
- Single webpage scraping: Extract content from any accessible webpage
- Content extraction: Use prompts to extract specific information
- Markdown output: Clean, structured content ready for AI processing
- Real-time scraping: Dynamic content extraction during AI conversations
Development
For development and testing:
git clone https://github.com/WebCrawlerAPI/webcrawlerapi-mcp
cd webcrawlerapi-mcp
npm install
npm run dev
Build for production:
npm run build
npm start
GitHub Repository
WebCrawlerAPI MCP Server on GitHub (opens in a new tab)
If you have any questions, please contact us at [email protected]