theboring.app
Back to projects

Insights

AI-powered GitHub activity analysis for developer growth and code review insights.

Why I made this

I manage a few developers and it takes me quite some time to review their work and performance.

I wanted a simple dashboard that I could spot performance trends and have a general overview of their work.

This dashboard is just one of the tools I use, it is not a replacement for a proper performance review.

Try it Live View Source

Project Details

30 days to build
133 commits
Released May 16, 2025

Tools & Technologies

Cursor Vercel Claude Code CLI

AI Models Used

claude-3.7-sonnet gemini-2.5-pro

Related Insights

Thoughts and learnings from building this project

I’m impressed with the Claude Code CLI, how intuitive and easy to use it is. However, super pricy for what it delivers.

I’ll keep working with Cursor as most of the time it’s enough.

Written with AI

A privacy-first developer analytics tool that leverages AI to analyze GitHub activity and provide actionable feedback for improving coding practices and developer growth.

Features

GitHub activity analysis: Comprehensive review of pull requests with detailed metrics, timeline visualization, and performance tracking across selected timeframes. Analyzes PR patterns, review cycles, and development velocity.

AI-powered code feedback: Individual pull request analysis using multiple AI providers (OpenAI, Anthropic, Google Gemini) to provide detailed code quality assessments, best practice recommendations, and improvement suggestions.

Pattern recognition system: Meta-analysis capabilities that identify recurring patterns across multiple PRs, helping developers recognize strengths, weaknesses, and growth opportunities over time.

Privacy-centric architecture: Completely browser-based application with no server-side code storage or processing. Direct GitHub API communication ensures sensitive code never leaves your environment.

Architecture

Built with React and TypeScript for type-safe development, featuring Zustand for efficient state management and Framer Motion for smooth animations. Recharts handles data visualization and analytics dashboard rendering.

Client-side processing: All analysis happens locally in the browser, with direct API calls to GitHub for repository data and user-configured AI providers for code analysis.

Multi-provider AI support: Flexible integration with OpenAI, Anthropic Claude, and Google Gemini APIs, allowing users to choose their preferred AI provider with their own API keys.

Technical implementation

Frontend: React with TypeScript, Tailwind CSS for styling, and Framer Motion for interactive animations and transitions.

State management: Zustand for lightweight, performant state handling across components without Redux complexity.

Data visualization: Recharts library for rendering pull request metrics, timeline charts, and developer growth analytics.

Security model: Browser-only architecture with user-provided API keys, ensuring complete data privacy and no third-party code storage.

The system demonstrates how to build sophisticated developer tools while maintaining strict privacy standards – all analysis happens locally with user-controlled AI provider access.