MCP Code Review Server

🚀 Code Review Server

A custom MCP server that uses Repomix and Large Language Models (LLMs) to perform code reviews.

🚀 Quick Start

This code review server is a custom MCP server that leverages Repomix and LLMs to conduct code reviews. It offers structured review results, supports multiple LLM providers, and can handle large codebases efficiently.

✨ Features

• Use Repomix to flatten code repositories.
• Analyze code using Large Language Models.
• Obtain structured code review results with specific issues and suggestions.
• Support multiple LLM providers (OpenAI, Anthropic, Gemini).
• Handle chunking of large code repositories.

📦 Installation

# Clone the repository
git clone https://github.com/yourusername/code-review-server.git
cd code-review-server

# Install dependencies
npm install

# Build the server
npm run build

📚 Documentation

Configuration

Create a .env file in the root directory based on the .env.example template:

cp .env.example .env

Edit the .env file to set your preferred LLM provider and API key:

# LLM provider configuration
LLM_PROVIDER=OPEN_AI
OPENAI_API_KEY=your_openai_api_key_here

Usage

As an MCP Server

This code review server implements the Model Context Protocol (MCP) and can be used with any MCP client:

# Start the server
node build/index.js

The server provides two main tools:

1. analyze_repo: Flatten the code repository using Repomix.
2. code_review: Perform code reviews using LLMs.

Usage Guide

analyze_repo

Use this tool when you need to:

• Get an overview of the code repository.
• Understand the purpose and structure of the code repository.
• Analyze the main components of the code repository.

Example usage:

# Example command
node build/index.js analyze_repo --path /path/to/your/repo

code_review

Use this tool when you need to:

• Conduct in - depth analysis of code quality.
• Identify potential issues and vulnerabilities.
• Get improvement suggestions.

Example usage:

# Example command
node build/index.js code_review --path /path/to/your/repo --focus SECURITY,MAINTAINABILITY

🔧 Technical Details

LLM Integration Details

1. Model Selection and Configuration:
   • By default, specific versions provided by the vendors are used, such as OpenAI's gpt - 4 - turbo, Anthropic's claude - 3 - sonnet, and Gemini's gemini - 1.5 - flash - preview.
   • Users can override the default model selection via environment variables.
2. API Calls and Error Handling:
   • Retry logic is implemented to enhance the stability of API calls.
   • Request timeouts and error handling mechanisms are included to ensure service reliability.
3. Code Processing Flow:
   • Use the Repomix tool to convert the code repository into linear text for easier model analysis.
   • Chunk the code according to the context window limit to ensure effective model processing.

Code Review Output Format

Code review results are returned in a structured JSON format:

{
  "summary": "A brief summary and purpose description of the code repository",
  "issues": [
    {
      "type": "SECURITY|PERFORMANCE|QUALITY|MAINTAINABILITY",
      "severity": "HIGH|MEDIUM|LOW",
      "description": "A specific description of the issue",
      "line_numbers": [12, 15],
      "recommendation": "A suggested fix"
    }
  ],
  "strengths": ["A list of the code repository's strengths"],
  "recommendations": ["Overall improvement suggestions"]
}

📄 License

MIT