r2r-mcp

r2r-mcp

0
knowledge_and_memoryai_chatbot

R2R MCP Server is a system designed to integrate Model Context Protocol with the R2R system, enhancing interactions with MCP-compatible models like Claude. It supports high-performance access to knowledge bases and features advanced search and response generation capabilities.

🚀 R2R MCP Server

R2R MCP Server provides integration of the Model Context Protocol (MCP) with the R2R (Retrieval-to-Response) system to enhance interaction with Claude and other MCP-compatible models, ensuring access to knowledge bases.

✨ Features

  • MCP server for integration with Claude and other MCP-compatible models
  • Asynchronous R2R client for high-performance access to knowledge bases
  • Tools for search and response generation based on contextual data
  • Structured logging with Loguru
  • Lightweight architecture with minimal dependencies
  • Advanced RAG (Retrieval-Augmented Generation) capabilities
  • Web search support for obtaining current information
  • Research capability through an agent for complex queries

🧰 Available MCP Tools

  • search: Document search using R2R, including vector and hybrid search
  • rag: Knowledge base response retrieval with source citation
  • web_search: Internet search through R2R with results from open sources
  • document_search: Search within specific documents or the entire knowledge base
  • list_documents: List available documents in the knowledge base with metadata
  • agent_research: Research through an agent for complex queries using LLM

🔍 Functionality

  • Vector and hybrid search for semantic similarity and precise matches
  • RAG for generating coherent responses based on documents
  • Agent-based research for deep analysis using powerful LLM models

🤖 Usage with Claude

Claude can access R2R tools for document search, question answering with citations, research by combining sources, and obtaining current information online.

👨‍💻 Development

Includes a Makefile for convenient task execution, local server launch, test execution, and code formatting and validation.

🧪 Testing

Comprehensive system testing: Unit tests for components, integration tests with real API calls, and performance tests evaluating component efficiency.

⚡ Performance

Optimized for high performance with asynchronous clients, efficient formatting, client caching, and strict performance tests.

🔄 Integration with Other MCP-Compatible Models

R2R MCP Server works with any MCP-compatible models, including Claude and other LLM models.