git-mcp-go MCP Server

What does “git-mcp-go” MCP Server do?

The git-mcp-go MCP Server is a Model Context Protocol (MCP) server designed for seamless interaction with Git repositories through Large Language Models (LLMs). By providing a standardized interface, it allows AI assistants and clients to read, search, and manipulate Git repositories programmatically. This enhances developer workflows by enabling tasks such as checking repository status, viewing diffs, staging changes, committing, creating branches, and pushing to remotes—all via conversational or automated LLM interfaces. The server supports operations across multiple repositories at once and offers both shell-based and Go-based implementations for flexibility and performance. This integration streamlines codebase management, makes automation more accessible, and empowers developers to leverage LLMs for complex Git-based operations.

List of Prompts

No explicit prompt templates are listed in the repository or documentation.

List of Resources

No explicit MCP resources are listed in the repository or documentation.

List of Tools

• git_status: Shows the working tree status of the repository.
• git_diff_unstaged: Displays changes in the working directory that have not yet been staged.
• git_diff_staged: Shows changes that are staged for the next commit.
• git_diff: Compares differences between branches or commits.
• git_commit: Records new changes to the repository with a commit.
• git_add: Adds file contents to the staging area for the next commit.
• git_reset: Unstages all staged changes.
• git_log: Displays the commit history (log) of the repository.
• git_create_branch: Creates a new branch, optionally from a specified base branch.
• git_checkout: Switches between different branches in the repository.
• git_show: Shows the contents of a specific commit.
• git_init: Initializes a new Git repository.
• git_push: Pushes local commits to a remote repository (requires --write-access).
• git_list_repositories: Lists all available Git repositories managed by the server.

Use Cases of this MCP Server

• Automated Codebase Management: Developers can use LLMs to automate tasks like checking status, diffing, and committing changes, reducing manual CLI work.
• Multi-Repository Operations: The server’s multi-repo support allows users to manage and query several repositories in a unified workflow.
• Branching and Collaboration: Easily create, checkout, and manage branches, enabling smoother collaboration and feature development workflows.
• Commit and History Exploration: AI models can help users review commit logs, inspect history, and understand code changes with natural language commands.
• Seamless Integration with AI Assistants: Enables integration with tools like Cline and others, allowing LLMs to operate as Git assistants within development environments.

How to set it up

Windsurf

1. Ensure Go (1.18+) and Git are installed on your system.
2. Download the prebuilt binary or build from source:

   git clone https://github.com/geropl/git-mcp-go.git
   cd git-mcp-go
   go build -o git-mcp-go .
   

   Or install directly:

   go install github.com/geropl/git-mcp-go@latest
   

3. Add the following JSON to your Windsurf MCP server configuration:

   {
     "mcpServers": {
       "git-mcp-go": {
         "command": "git-mcp-go",
         "args": ["serve", "-v", "/path/to/repo1", "/path/to/repo2"]
       }
     }
   }
   

4. Save the configuration and restart Windsurf.
5. Verify the setup by running a test Git command via the MCP integration.

Claude

1. Prepare your system with Go (1.18+) and Git.
2. Download or build the git-mcp-go binary as above.
3. Edit your Claude MCP configuration file to include:

   {
     "mcpServers": {
       "git-mcp-go": {
         "command": "git-mcp-go",
         "args": ["serve", "--mode", "go-git", "-r", "/path/to/repo"]
       }
     }
   }
   

4. Restart Claude’s backend service.
5. Test by issuing a Git status or log request through Claude.

Cursor

1. Install Go and Git if not already present.
2. Obtain the git-mcp-go binary, via build or prebuilt.
3. Update Cursor’s configuration file:

   {
     "mcpServers": {
       "git-mcp-go": {
         "command": "git-mcp-go",
         "args": ["serve", "-v", "/path/to/repo1"]
       }
     }
   }
   

4. Save and restart Cursor.
5. Confirm integration through Cursor’s MCP server diagnostics.

Cline

1. Ensure Go and Git are available.
2. Build or download the git-mcp-go binary.
3. Use the setup command for Cline:

   ./git-mcp-go setup -r /path/to/git/repository
   

4. Alternatively, configure cline_mcp_settings.json:

   {
     "mcpServers": {
       "git-mcp-go": {
         "command": "git-mcp-go",
         "args": ["serve", "-v", "/path/to/repo"]
       }
     }
   }
   

5. Restart Cline and check the status from within the app.

Securing API Keys

If you need to secure sensitive information (e.g., API keys), use environment variables:

{
  "mcpServers": {
    "git-mcp-go": {
      "command": "git-mcp-go",
      "args": ["serve", "-v", "/path/to/repo"],
      "env": {
        "GIT_MCP_KEY": "${GIT_MCP_KEY}"
      },
      "inputs": {
        "some_key": "${SOME_KEY}"
      }
    }
  }
}

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "git-mcp-go": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “git-mcp-go” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Based on the tables above, the git-mcp-go MCP server offers a robust toolset for Git automation with LLMs, clear setup instructions, and strong multi-platform support, but lacks explicit resource exposure, prompt templates, and advanced MCP features like roots or sampling.

MCP Score

Rating:
I would rate this MCP server a 6/10. It is functionally strong for Git-based operations and is well-documented for setup, but lacks MCP prompt/resource richness, a license, and support for advanced MCP features (roots, sampling). Its practical utility for developers is high, but completeness as an MCP integration could be improved.