Docs Search Engine

关于

This custom Model Context Protocol (MCP) server functions as a personalized documentation search engine, inspired by Context7. It empowers AI assistants by providing access to the latest information from GitHub repositories and web pages, ensuring your AI is grounded in current reality rather than outdated data. By ingesting, indexing, and caching documentation, it delivers relevant context directly to your LLM-based tools, enhancing their accuracy and utility.

主要功能

• Real-time documentation ingestion from GitHub repositories and web pages.
• Efficient in-memory full-text search powered by minsearch.
• 0 GitHub stars
• Web scraping functionality using Jina Reader to convert web content into LLM-friendly markdown.
• Seamless integration with AI assistants via the Model Context Protocol (FastMCP framework).
• URL-based caching of indexed content for rapid subsequent searches.

使用案例

• Scrape and summarize content from any specified web page.
• Enable AI assistants to search for specific information within FastMCP or minsearch documentation.
• Count word occurrences on web pages to analyze text content with an AI assistant.