MCP Server to Auto commit changes 🛠️
This implementation provides a Git changes analyzer that generates commit messages using OpenAI’s GPT models.
Demo

Features
- Analyzes git changes in your repository (both staged and unstaged)
- Generates conventional commit messages using GPT-4o-mini
- Provides detailed summaries of:
- 📝 Modified files
- ✨ Newly added files
- 🗑️ Deleted files
- 📄 Detailed changes (up to 10 lines per file)
- Built with @modelcontextprotocol/sdk
- Adds an auto-commit signature to each commit
Project Structure
mcp-server-auto-commit/
├── index.ts # Main server implementation with git analysis tool
├── package.json # Project dependencies
├── tsconfig.json # TypeScript configuration
└── build/ # Compiled JavaScript output
Prerequisites
- Node.js installed
- OpenAI API key
- Git repository to analyze
- pnpm package manager
Getting Started
- Clone this repository:
git clone https://github.com/jatinsandilya/mcp-server-auto-commit.git
cd mcp-server-auto-commit
- Install dependencies:
pnpm install
Set up your OpenAI API key using one of these methods:
- Set as an environment variable:
OPENAI_API_KEY=your-api-key - Pass as a command line argument:
--key your-api-key - Add to a
.envfile in the project root
- Set as an environment variable:
Build the project:
pnpm run build
This will generate the /build/index.js file - your compiled MCP server script.
Using with Cursor
- Go to Cursor Settings -> MCP -> Add new MCP server
- Configure your MCP:
- Name: git-auto-commit
- Type: command
- Command:
node ABSOLUTE_PATH_TO_MCP_SERVER/build/index.js --key your-api-key(Replaceyour-api-keywith your actual OpenAI API key if not set in environment)
Using with Claude Desktop
Add the following MCP config to your Claude Desktop configuration:
{
"mcpServers": {
"git-auto-commit": {
"command": "node",
"args": ["ABSOLUTE_PATH_TO_MCP_SERVER/build/index.js", "--key", "your-api-key"]
}
}
}
Available Tools
git-changes-commit-message
This tool analyzes the current git changes in your repository and generates a commit message using OpenAI’s GPT-4o-mini model. It provides:
- List of modified files with status indicators
- List of newly added files
- List of deleted files
- Detailed changes for each file (limited to 10 lines per file for readability)
- A generated commit message following conventional commits format
- An auto-commit signature
Usage parameters:
autoCommitPath: Optional path to analyze specific directory/file. If not provided, uses current working directory.
Development
The implementation in index.ts showcases:
- Setting up the MCP server with proper configuration
- Handling command line arguments and environment variables
- Integrating with OpenAI’s API using GPT-4o-mini model
- Git operations using child processes
- Error handling and fallback mechanisms
- Detailed change analysis and formatting
To modify or extend the implementation:
- Update the server configuration in
index.ts:
const server = new McpServer({
name: "git-auto-commit",
version: "0.0.1",
});
The tool is defined using
server.tool()with proper parameter validation using Zod schema.Build and test your changes:
pnpm run build
Contributing
Feel free to submit issues and enhancement requests!
License
MIT
Git Auto Commit
Project Details
- jatinsandilya/mcp-server-auto-commit
- send-email
- MIT License
- Last Updated: 4/1/2025
Categories
Recomended MCP Servers
LongPort OpenAPI SDK Base.
A Model Context Protocol server implementation for ClickUp integration, enabling AI assistants to interact with ClickUp workspaces.
Node.js/TypeScript MCP server for Atlassian Bitbucket. Enables AI systems (LLMs) to interact with workspaces, repositories, and pull requests...
MCP server for aiding with literature reviews
This is a TypeScript-based Model Context Protocol (MCP) server that implements a virtual pet simulation system. It demonstrates...
Memory Bank is an MCP server that helps teams create, manage, and access structured project documentation. It generates...
Payload CMS MCP Server
MCP Think Tool Claude Desktop
Cryptocurrency Market Data MCP Server
MCP Python Interpreter: run python code. Python-mcp-server, mcp-python-server, Code Executor
MCP server for Grafana





