Track and measure
AI-assisted development

AIDA Metrics provides tangible, auditable metrics to distinguish between AI noise and AI value in your software development workflow.

npm install -g @aida-dev/cli
View on GitHub
Terminal
$ aida collect --since 30d
✓ Collected 247 commits from last 30 days
✓ AI-tagged commits: 89 (Claude, Copilot, ChatGPT)
$ aida analyze
✓ Merge ratio: 78.5%
✓ Average persistence: 12.3 days
$ aida report
✓ Reports generated in ./aida-output/
$ aida comment
✓ AIDA report posted to PR #42

Core Metrics

📊

Merge Ratio

Track what percentage of AI-generated code actually gets merged into production.

⏱️

Persistence

Measure how long AI-generated code survives in your codebase before being modified.

🎯

AI Detection

4-level attribution classification (explicit, implicit, mention, none) for Claude Code, Copilot, ChatGPT, Cursor, Windsurf, Gemini, Codeium and more.

💬

PR Comments

Post AIDA reports directly on pull requests. Auto-detects GitHub Actions with GitLab and Bitbucket coming soon.

🚀

CI/CD Ready

One workflow file, zero configuration. Runs on every PR and updates the comment automatically.

Get Started

1

Install globally

npm install -g @aida-dev/cli
2

Navigate to your Git repo

cd /path/to/your/repo
3

Run analysis

aida collect --since 30d && aida analyze && aida report

Supported AI Tools

AIDA classifies commits into explicit, implicit, and mention levels via commit messages, git trailers, and Co-Authored-By attribution.

Claude Code
GitHub Copilot
ChatGPT
Cursor
Windsurf
Gemini
Codeium
+ Custom tools via .aida.json

CI/CD Integration

Add one workflow file to your repo and get AI metrics on every pull request.

.github/workflows/aida.yml
name: AIDA Analysis
on: [pull_request]
 
jobs:
  aida:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
      - run: npm i -g @aida-dev/cli
      - run: aida collect --pr && aida analyze && aida report && aida comment
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

Ready to measure AI impact?

Join developers who are quantifying their AI-assisted development workflow.