GitHub Copilot Usage Tracker
📊 Understand your GitHub Copilot usage — Track tokens, analyze tools, and gain actionable insights into your AI-powered development workflow.
Get real-time visibility into your Copilot usage with beautiful dashboards, intelligent request categorization, and comprehensive analytics — all stored locally and completely private.

✨ Key Features
📈 Real-Time Usage Metrics
Track every Copilot interaction with comprehensive metrics displayed on an intuitive dashboard:
- Token consumption by model and time period
- Request counts with success/failure tracking
- Session analytics across all your projects
- Live status bar showing current usage stats
📊 Rich Data Visualizations
Powerful charts that make your usage patterns clear at a glance:

- Model distribution — See which Copilot models you use most
- Usage by category — Understand what you're asking Copilot for
- Daily trends — Track token consumption over time
- Project breakdowns — Compare usage across workspaces
Monitor which Copilot tools are being used and how they perform:

- Success/failure rates for every tool
- Execution time tracking to identify bottlenecks
- Usage frequency to see your most-used tools
- Visual progress indicators for quick insights
🏷️ Smart Auto-Categorization
Every Copilot request is automatically classified into categories like:
- Development, Testing, Debugging
- Refactoring, Documentation
- Architecture, Optimization
- And more — fully customizable!
🚀 Getting Started
- Install the extension from the VS Code Marketplace
- Use GitHub Copilot as you normally would
- Click the status bar item (bottom right) to open your dashboard
- Explore your insights — filtering by time, project, or category
That's it! The extension works automatically in the background, requiring zero configuration.
💡 Why You'll Love It
✅ Zero Configuration Required
Start tracking immediately — no setup, no API keys, no external services.
✅ Complete Privacy
All data stays on your machine. No cloud uploads, no external calls, no tracking.
✅ Flexible Time Ranges
View your stats for:
- Today — See real-time activity
- 7 days — Weekly patterns (default)
- 30 days — Monthly overview
- All time — Complete history
✅ Multi-Project Support
Track usage across all your workspaces with per-project filtering and comparison.
✅ Fully Customizable
- Define your own categorization taxonomy
- Choose which AI model powers the labeling
- Set custom storage locations
- Auto-purge old data to save space
📋 Commands
Access via Command Palette (Cmd+Shift+P / Ctrl+Shift+P):
GHCP Tracker: Show Usage Statistics — Open the dashboard
GHCP Tracker: Export Statistics — Save data as JSON
GHCP Tracker: Reset Statistics — Clear all data
GHCP Tracker: Open Data Location — View session files
GHCP Tracker: Purge Old Session Files — Clean up old data
⚙️ Configuration (Optional)
Customize the extension in VS Code Settings:
{
"ghcp-tracker.enableLabeling": true,
"ghcp-tracker.labelingModel": "gpt-4o-mini",
"ghcp-tracker.trackTokens": true,
"ghcp-tracker.trackTools": true,
"ghcp-tracker.autoPurgeDays": 30
}
Key Settings:
enableTracking — Enable/disable tracking (default: true)
enableLabeling — Auto-categorize requests (default: true)
labelingModel — AI model for labeling (default: "gpt-4o")
labelCategories — Custom taxonomy (default: 10 categories)
autoPurgeDays — Auto-delete old sessions (default: 30)
View all configuration options →
🎯 Perfect For
Individual Developers — Understand your coding patterns and optimize Copilot usage
Team Leads — Monitor team adoption and identify training opportunities
Organizations — Track ROI, manage budgets, and optimize AI investments
Researchers — Analyze AI-assisted development behaviors with exportable data
🔒 Privacy & Security
- ✅ 100% local — All data stored on your machine
- ✅ No telemetry — We don't collect or transmit anything
- ✅ Open source — Audit the code yourself
- ✅ You control the data — Export, delete, or move it anytime
📚 Learn More
📄 License
MIT © cykruss
Love this extension? Please ⭐ star the repository and leave a review on the marketplace!
Questions or feedback? Open an issue — we'd love to hear from you.