RepoHealth-AI: Made using Github Copilot CLI

This is a submission for the GitHub Copilot CLI Challenge

What I Built

RepoHealth AI is a full-stack web application that analyzes any public GitHub repository and gives it a comprehensive health score powered by real data and AI-generated insights.

The idea came from a simple frustration — when you discover a new open source repo, it's hard to quickly judge whether it's well-maintained, properly documented, or worth depending on. RepoHealth AI solves that in seconds.

Here's what it does:

You paste any GitHub repository URL and the app instantly fetches live data from the GitHub API and displays a full dashboard including:

• 
  Repository Health Score — A calculated score out of 100 based on README quality, license presence, commit frequency, issue activity, and contributor count


• 
  Key Stats — Stars, forks, open/closed issues, watchers, contributors, last commit date, and primary language


• 
  Commit Activity Chart — Visual representation of commits over the last 30 days


• 
  Language Breakdown — Pie chart showing language distribution with official language colors


• 
  Top Contributors — Bar chart of the most active contributors by commit count


• 
  Issues Overview — Visual breakdown of open vs closed issues


• 
  AI-Powered Insights — On-demand AI analysis (powered by Google Gemini) that provides a repo summary, key strengths, and prioritized improvement suggestions with High/Medium/Low priority labels

The app features a full dark/light theme toggle, responsive design across all screen sizes, smooth animations, and a clean professional UI built to feel like a real developer tool.

Live App: https://repohealth-ai.vercel.app

GitHub Repo: https://github.com/Muhammad-Ahmed-Rayyan/RepoHealth-AI

Demo

🔗 Live Demo: https://repohealth-ai.vercel.app

How to test:

1. Visit the live app


2. Paste any public GitHub repository URL (e.g., https://github.com/facebook/react)


3. Hit the search button and wait for the dashboard to load


4. Scroll down and click "Generate AI Insights" to get AI-powered recommendations

Screenshots:

Landing Page — Clean search interface with dark theme

Dashboard — Full repository health analysis with charts and stats

My Experience with GitHub Copilot CLI

GitHub Copilot CLI was not just a helper in this project — it was essential to the entire build. From the very first line of code to the final deployment, every part of this project was built using Copilot CLI directly in the terminal.

How I used it:

The entire project architecture was scaffolded using Copilot CLI. I described what I wanted — a full-stack app with a React frontend, Node.js/Express backend, GitHub API integration, and AI-powered analysis — and Copilot CLI generated the complete project structure, boilerplate, and initial implementations.

Where it made the biggest difference — Debugging:

The most impressive part of working with Copilot CLI was how fast it helped resolve bugs. For example:

• When the GitHub API was returning 401 Unauthorized errors, Copilot CLI diagnosed the root cause immediately — the process.env.GITHUB_TOKEN was being read at module import time before dotenv.config() had run. It suggested converting the static axios instance into a factory function that reads the token at runtime. Problem solved in minutes.




• When the AI analysis was failing with "Cannot read properties of undefined (reading 'exists')", Copilot CLI traced the issue to a data flow problem — the repo data wasn't being passed correctly to the AI service after separating AI generation into its own triggered flow. It restructured the API call chain and added optional chaining (?.) throughout the AI service.




• When Google Gemini returned a 404 model not found error, Copilot CLI suggested running a curl command to list available models from the API, then immediately updated the model name to the correct one.

Other highlights:

• Generated all Recharts chart components with proper dark theme configurations


• Built the health score calculation algorithm from scratch based on my requirements


• Helped design the responsive Tailwind CSS layout and card system


• Implemented the dark/light theme toggle with localStorage persistence


• Wrote the complete prompt engineering for the Gemini AI analysis section

Working with Copilot CLI felt like having a senior developer pair programming with me directly in the terminal. The speed at which it could read existing code, understand context, and apply targeted fixes was genuinely impressive. Tasks that would have taken hours of Stack Overflow searching were resolved in a matter of minutes.

This project genuinely would not have been completed in the available time without Copilot CLI.

Team Members: @ahmed_waseem_ec0fb5a03620 and @waleed_zaidi_92fd4c9733df

GitHub Copilot CLI in Action