LLM-MCP-RAG Python实现

LLM-MCP-RAG-py

Dependency

pip install -r requirement.txt

Also need to install nodejs

How To Run

• get chat api from openrouter

• get embedding api from siliconflow

• create and put api key in .env file

  example of .env

  复制

  OPENAI_API_KEY=xxx
  OPENAI_BASE_URL=https://openrouter.ai/api/v1
  
  EMBEDDING_KEY=xxx
  EMBEDDING_BASE_URL=https://api.siliconflow.cn/v1
  

• To start the main program, use the following command:

  复制

  python ./src/main.py
  

  To test the chat functionality module, run:

  复制

  python ./src/test_chat.py
  

  To test the MCP client functionality, execute:

  复制

  python ./src/testMCPClient.py
  

  To test the Agent module functionality, run:

  复制

  python ./src/Agent.py
  

  Note: When testing the Agent, make sure to set up the MCP client correctly. Modify the directory path r'C:\Users\32114\Desktop\code\llm-mcp-rag-py\output' to suit your needs, allowing access to the appropriate directory.

LLM

• OpenAI API

MCP

• MCP 架构
• MCP Client
• Fetch MCP
• Filesystem MCP

RAG

• Retrieval Augmented Generation

  • 译文: https://www.yuque.com/serviceup/misc/cn-retrieval-augmented-generation-overview

• 硅基流动

• json数据

Acknowledgments

This project is based on the work of llm-mcp-rag, which was originally implemented in TypeScript. This implementation has been rewritten using Python