MCP Code Analyzer Setup & Run Instructions

This project demonstrates an MCP (Model Context Protocol) server that integrates an open-source LLM (GPT-J 6B) to analyze a GitHub repository’s code and answer questions about it.

Prerequisites

• Python 3.10 or higher installed on your system.
• Git to clone the repository.

Step 1: Clone the Repository

git clone https://github.com/your-username/your-repo-name.git
cd your-repo-name
  

Step 2: Create and Activate a Virtual Environment

It’s recommended to use a virtual environment for this project:

# Windows:
python -m venv venv
venv\Scripts\activate
macOS/Linux:
python3 -m venv venv
source venv/bin/activate

Step 3: Install Dependencies

Install all required packages using the provided requirements.txt file:

pip install -r requirements.txt
  

Step 4: Run the MCP Server

In one terminal, start the MCP server:

# Example using the default open-source model:
python mcp_server.py
To use ChatGPT or Claude, set the following environment variables first:
LLM_CHOICE=chatgpt  # or "claude"
OPENAI_API_KEY=your-openai-key      # required for ChatGPT
ANTHROPIC_API_KEY=your-anthropic-key  # required for Claude
python mcp_server.py

You should see a message indicating that the server has started (e.g., “Starting MCP server on port 5000...”).

Step 5: Run the MCP Client

Open another terminal (or tab) with the same virtual environment activated, then run:

python mcp_client.py
  

The client will connect to the server, analyze the GitHub repo, and query the LLM for an answer.

Troubleshooting

• Python Version Errors: Make sure you’re on Python 3.10 or higher.
• Virtual Environment Issues: Double-check that your environment is activated before installing or running.
• LLM Loading Problems: The default CodeLlama model may require significant resources. You can switch models by setting LLM_CHOICE to chatgpt or claude if you have the appropriate API keys.

© 2023 Your Name or Organization