R2r MCP

🚀 R2R MCP Server

R2R MCP Server is a powerful tool for interacting with LLMs. It supports integrating multiple models through the MCP protocol and offers a rich set of functional modules.

🚀 Quick Start

R2R MCP Server provides a convenient way to interact with LLMs. It supports various models via the MCP protocol and offers a wide range of features.

✨ Features

Model Interaction

• Support conversations with various LLMs via the MCP protocol.
• Provide context management to maintain conversation history.
• Support file attachment upload and download.

Functional Modules

• Document Parsing: Automatically parse documents in formats such as PDF and Word.
• Code Interpretation: Understand and explain code in multiple programming languages.
• Data Processing: Support the processing and analysis of data in formats like CSV and JSON.
• Image Recognition: Integrate OCR technology to support image content recognition.

Performance Optimization

• Support asynchronous request processing.
• Provide a result caching mechanism.
• Ensure system stability through rigorous performance testing.

📦 Installation

Install Dependencies

pip install mcp r2r loguru pytest pytest-asyncio pytest-cov black isort mypy ruff httpx

Initialize the Project

mcp init --name "R2R MCP Server" --description "A powerful LLM interaction server"

📚 Documentation

Environment Variables

• R2R_API_KEY: The access key for the R2R service.
• MCP_SERVER_ADDRESS: The address of the MCP server.
• LOG_LEVEL: The log level (DEBUG, INFO, WARNING, ERROR).

Log Configuration

import logging

logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
)

Development Guide

Function Extension

1. Add new model support: Implement the ModelAdapter interface to adapt to specific models.
2. Extend functional modules: Add new functional plugins in the extensions directory.
3. Customize command-line tools: Implement custom commands through cli.py.

Contribution Guide

1. Fork the repository.
2. Create a feature branch: git checkout -b feature/amazing-feature.
3. Commit your code and merge it into the main branch.

Performance Optimization

• Use asynchronous processing to improve system response speed.
• Integrate a caching mechanism to reduce redundant calculations.
• Regularly clean up invalid contexts to release resources.

Integrate Other MCP-Compatible Models

R2R MCP Server supports integration with various MCP-compatible models, including:

• Claude (via Claude.app or API)
• Other LLM models
• Custom MCP clients

Dependency Management

The main dependencies of the project are as follows:

Project Structure

r2r-mcp-server/
├── server.py          # Main service program
├── extensions/       # Functional extension modules
│   ├── document_parsing.py    # Document parsing module
│   └── code_interpreter.py     # Code interpreter module
├── cli.py            # Command-line tool
└── tests/           # Test cases
    └── test_server.py      # Service test cases

💻 Usage Examples

Basic Usage

from r2r_mcp_server import MCPClient

client = MCPClient("http://localhost:8000")
response = client.chat_completion(
    messages=[{"role": "user", "content": "你好，我是测试用户。"}]
)
print(response)

📄 License

MIT License

Project Maintainers

• Zhang San (peter.zhang@example.com)
• Li Si (jack lee@example.com)