Keyword RAG icon

Keyword RAG

Enables efficient knowledge retrieval from markdown documents using a BM25 algorithm.

About

This tool provides a Model Context Protocol (MCP) server designed for fast and accurate knowledge retrieval from a collection of markdown-based documents. It leverages the BM25 algorithm to perform keyword-based searches across structured document domains, making it ideal for creating searchable internal knowledge bases or customer support systems. The server is highly configurable, allowing users to define document sources, optimize search parameters, and seamlessly integrate with AI assistants like Claude Desktop for enhanced contextual responses.

Key Features

  • Configurable document domains for organized knowledge retrieval
  • Seamless integration with Claude Desktop via MCP
  • Supports Markdown (.md) and MDX (.mdx) document formats
  • Provides API tools for document search, retrieval, and domain listing
  • BM25-based document search for relevant results
  • 0 GitHub stars

Use Cases

  • Providing quick and contextual answers for customer support
  • Building an internal knowledge base for company information
  • Integrating existing documentation into AI assistant workflows
Keyword RAG: BM25 Knowledge Retrieval for Markdown Docs