MCP Experts

🚀 MCP Code Expert System

A Python-based code review system using the Model Context Protocol (MCP). It provides code review capabilities by simulating expert roles such as Martin Fowler and Robert C. Martin (Uncle Bob).

🚀 Quick Start

This MCP Code Expert System is a Python - based code review system. It leverages the Model Context Protocol (MCP) and simulates expert roles to offer comprehensive code review services.

✨ Features

• Code review based on Martin Fowler's refactoring principles.
• Code review based on Robert C. Martin's Clean Code principles.
• Knowledge graph to store code, comments, and their correlations.
• Integration with Ollama for AI - driven reviews.
• Support for Server - Sent Events (SSE) for web integration.

📦 Installation

Prerequisites

Python 3.10+

This project requires Python 3.10 or a higher version.

Ollama

Ollama is used for AI - driven code reviews.

1. Install Ollama according to your platform:
   • macOS: Download from ollama.com.
   • Linux: curl -fsSL https://ollama.com/install.sh | sh
   • Windows: Use Linux commands supported by WSL2.
2. Pull the recommended model:

   ollama pull llama3:8b
   

3. Start the Ollama server:

   ollama serve
   

Install Dependencies

Run the installation script to install dependencies and create a virtual environment:

chmod +x setup.sh
./setup.sh

📚 Documentation

Configuration

Edit the .env file for configuration (create it from .env.example if needed):

# Knowledge graph settings
KNOWLEDGE_GRAPH_PATH=data/knowledge_graph.json

# Ollama configuration (local AI model)
OLLAMA_HOST=http://localhost:11434
OLLAMA_MODEL=llama3:8b

Usage

Run the Server

Standard Mode (for Cursor integration)

source .venv/bin/activate
python server.py

HTTP/SSE Mode (for Web integration)

source .venv/bin/activate
python server.py --transport sse

This will start a server with SSE transport at http://localhost:8000/sse.

Customize the port:

python server.py --transport sse --port 9000

Install in Cursor

To install in the Cursor IDE:

source .venv/bin/activate
mcp install server.py --name "Code Expert System"

Available Tools

The server exposes the following tools:

• ask_martin: Ask Martin Fowler to review the code and suggest refactoring.
• ask_bob: Ask Robert C. Martin (Uncle Bob) to review the code according to Clean Code principles.
• read_graph: Read the entire knowledge graph.
• search_nodes: Search for nodes in the knowledge graph.
• open_nodes: Open specific nodes by name.

Example Usage

Use Martin Fowler to review a code snippet:

{
  "code": "function calculateTotal(items) {\n  var total = 0;\n  for (var i = 0; i < items.length; i++) {\n    total += items[i].price;\n  }\n  return total;\n}",
  "language": "javascript"
}

Project Structure

• server.py: The main server file.
• config.py: Configuration settings.
• knowledge_graph.py: Knowledge graph functionality.
• models/: AI model interfaces.
• templates/: Web interface templates.

Architecture

The system consists of the following modules:

• Server: Handles requests and coordinates components.
• Knowledge Graph: Stores and manages code and its relationships.
• Ollama Integration: Interacts with AI services for intelligent reviews.
• Web Interface: A user - friendly interface for code reviews.

📄 License

MIT License