home / mcp / git metrics mcp server

Git Metrics MCP Server

Provides access to git repository metrics, team health signals, and collaboration insights via MCP client queries.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "jonmatum-git-metrics-mcp-server": {
      "command": "git-metrics-mcp-server",
      "args": [],
      "env": {
        "GIT_TIMEOUT": "30s"
      }
    }
  }
}

You can run the Git Metrics MCP Server to analyze git repository metrics and health signals for your team. It exposes a set of data extraction tools that let you answer questions about commits, authors, file churn, team velocity, and code ownership, all through natural language queries from an MCP client.

How to use

Use an MCP client to connect to the Git Metrics MCP Server and ask questions in natural language. You can start with high-level team insights and then drill into specifics like burnouts, file churn, or conventional commit adoption. Common workflows include checking overall team health over a time period, identifying potential bus factor risks, and exploring collaboration patterns across repositories.

How to install

Prerequisites: make sure you have Node.js and npm installed on your system. You should have a working shell environment where you can run commands.

# Option 1: install globally (recommended for quick start)
npm install -g @jonmatum/git-metrics-mcp-server

# Option 2: install from source
git clone https://github.com/jonmatum/git-metrics-mcp-server.git
cd git-metrics-mcp-server
npm install
npm run build
```}]} ,{

Configuration and deployment notes

You can run the MCP server in several ways depending on how you prefer to install it. The following configurations are provided for typical usage patterns. Each configuration assumes you will connect to the server via an MCP client.

Environment variable you may see in practice: GIT_TIMEOUT. This controls the timeout for git operations. Default is 30 seconds if not set.

MCP server connection configurations

{
  "mcpServers": {
    "git_metrics": {
      "type": "stdio",
      "command": "git-metrics-mcp-server",
      "args": []
    }
  }
}

{
  "mcpServers": {
    "git_metrics": {
      "type": "stdio",
      "command": "npx",
      "args": ["@jonmatum/git-metrics-mcp-server"]
    }
  }
}

{
  "mcpServers": {
    "git_metrics": {
      "type": "stdio",
      "command": "npx",
      "args": ["tsx", "/path/to/git-metrics-mcp-server/src/git-metrics.ts"]
    }
  }
}

Notes on usage with Kiro CLI

Open the Kiro CLI and enter chat mode. Then ask questions like: “Get commit stats for /path/to/repo since 2025-11-01” or “Show me team velocity trends for the last 2 weeks.” The MCP server will fetch and summarize metrics from the target repository and present actionable insights.

Security and reliability notes

The server includes input sanitization and structured logging to help protect against injection attacks and to make monitoring easier. Errors are handled gracefully with detailed logs so you can diagnose issues without dropping data.

Development and testing

Run the development and test commands to validate your setup during integration.

npm run dev
npm run build
npm start

Available tools

get_commit_stats

Get overall commit statistics for a given time period including total commits, additions, deletions, files changed, and net change.

get_author_metrics

Detailed metrics per contributor such as commits, additions, deletions, and files touched.

get_file_churn

Identify files with the most changes over a period to highlight complexity or risk.

get_team_summary

Comprehensive summary of team activity and contribution across a period.

get_commit_patterns

Analyze commit timing patterns to detect burnout signals and work distribution.

get_code_ownership

Assess knowledge distribution and identify bus factor risks based on file ownership.

get_velocity_trends

Track velocity over time to observe productivity and pacing.

get_collaboration_metrics

Measure team interaction patterns and collaboration signals.

get_quality_metrics

Provide indicators of code quality such as churn, commit size, reverts, and fixes.

get_technical_debt

Identify stale files and hotspots of complexity that may indicate technical debt.

get_conventional_commits

Analyze usage of conventional commits, scopes, and release patterns.