Fetcher MCP

Fetcher MCP is an MCP server for fetching and extracting web page content using Playwright's headless browser capabilities. It is designed to extract clean, readable content from JavaScript-heavy or dynamic websites and output it in HTML or Markdown format.

Features

• JavaScript Rendering: Uses Playwright to execute JavaScript, enabling scraping of modern, dynamic web applications.
• Intelligent Content Extraction: Built-in Readability algorithm automatically extracts the main content, removing ads, navigation, and other non-essential elements.
• Flexible Output Formats: Supports both HTML and Markdown output formats.
• Parallel Processing: The fetch_urls tool allows concurrent fetching of multiple URLs for efficient batch operations.
• Resource Optimization: Automatically blocks unnecessary resources (images, stylesheets, fonts, media) to reduce bandwidth usage and improve performance.
• Robust Error Handling: Comprehensive error handling and logging for reliable operation.
• Configurable Parameters: Fine-grained control over timeouts, content extraction, and output formatting. Key parameters include:
  • url or urls: URL(s) to fetch
  • timeout: Page loading timeout (default 30s)
  • waitUntil: Navigation completion trigger (options: 'load', 'domcontentloaded', etc.)
  • extractContent: Whether to extract main content (default true)
  • maxLength: Maximum length of returned content
  • returnHtml: Return HTML instead of Markdown (default false)
  • waitForNavigation: Wait for additional navigation (for anti-bot, CAPTCHAs)
  • navigationTimeout: Additional navigation timeout (default 10s)
  • disableMedia: Disable loading of media resources (default true)
  • debug: Enable debug mode to show browser window
• Batch Operations: fetch_urls supports parallel fetching of multiple URLs with combined results.
• Debug Mode: Option to run with browser window visible for debugging or manual authentication.
• Custom Authentication: Supports manual login via debug mode, allowing scraping of authenticated pages.
• Development Tools: Includes scripts for installing dependencies, browsers, building the server, and debugging.

Usage

• Run directly with npx -y fetcher-mcp.
• Debug mode available with --debug flag.
• Can be configured as an MCP server for integration with other tools (e.g., Claude Desktop).

Licensing

• Licensed under the MIT License.

Source

• GitHub Repository

Pricing

• Fetcher MCP is open source and available for free under the MIT License.