Introducing Codemetry

Open Source

I’m excited to introduce Codemetry, a new open-source project I’ve been working on.

What is Codemetry?

Codemetry is a Git repository analysis tool that converts your version control history into meaningful quality metrics. It generates a daily quality indicator (good/medium/bad) that helps teams understand code quality trends through measurable data rather than guesswork.

How it works

Codemetry examines your Git history and detects patterns that often correlate with code quality:

  • Code churn: The volume of changes in a given period
  • File scatter: How modifications are distributed across files
  • Follow-up fixes: Rapid patches after commits, which can indicate rushed or problematic work
  • Commit patterns: Signals about development workflow

Each day’s metrics are compared against your project’s historical baseline. This means “high churn” is contextual to your specific repository, not an arbitrary threshold.

Key features

  • Baseline comparison for normalised insights specific to your project
  • Follow-up fix detection to identify potentially rushed work
  • Optional AI summaries that explain what the metrics mean
  • Laravel-first with native Artisan command support
  • Privacy-focused: Runs entirely locally with no code uploads; AI features are opt-in

Benefits

Codemetry helps you identify periods in your codebase that might need closer review. Instead of relying on gut feeling, you get data-driven insights. It’s particularly useful for:

  • Spotting risky periods before they become technical debt
  • Understanding the rhythm of your development process
  • Making informed decisions about when to schedule code reviews

Built with AI

Codemetry was fully implemented using an AI coding agent, based on my original idea and requirements. It’s a practical example of how AI can accelerate the development of useful tools when guided by clear human intent.

Get started

Check out the GitHub repository and the documentation to get started.

composer require albertoarena/codemetry
php artisan codemetry:analyze --days=7

I’d love to hear your feedback!

Alberto Arena

Alberto Arena

Senior Software Engineer