FetchV2

关于

This Model Context Protocol (MCP) server is engineered to streamline web content acquisition and extraction, primarily for AI agents. It leverages the Trafilatura library to intelligently fetch webpages, remove boilerplate elements like navigation bars, advertisements, and footers, and deliver pristine markdown output. Beyond basic content fetching, FetchV2 supports batch processing for retrieving content from multiple URLs in a single request and offers robust link discovery capabilities with filtering, ensuring AI models receive high-quality, relevant information efficiently.

主要功能

• Complies with robots.txt rules and includes graceful timeout handling.
• Offers pagination support for large pages using a start index.
• Extracts clean markdown content from any URL, removing boilerplate.
• Discovers and filters links on webpages for targeted exploration.
• Supports batch fetching of up to 10 URLs in a single request.
• 0 GitHub stars

使用案例

• Automatically fetch documentation or articles for AI agent analysis.
• Discover and filter related links on a webpage for targeted research.
• Summarize differences across multiple web pages by batch fetching their content.