RepoGuide AI

• 

  Complexity Score

• 

  Language analysis

• 

  Installation Guide

• 

  Project Sumamry

• 

  Architecture Diagram

• 

  Landing Page

Inspiration

Every developer wastes hours trying to understand a new codebase. I wanted to fix that with AI.

What it does

RepoGuide AI takes any GitHub repository URL and instantly generates a complete onboarding guide — including project overview, setup instructions, tech stack detection, language breakdown chart, code complexity score, interactive file tree, architecture diagram, and a step-by-step installation wizard.

How we built it

Built with React + Vite + Tailwind CSS for the frontend. Used the GitHub REST API to fetch repo data and Groq AI (Llama 3.3 70B) to analyze and generate all guide content. Charts built with Chart.js and diagrams with custom SVG.

Challenges we ran into

Getting Groq to return consistent JSON responses required careful prompt engineering. Rendering the architecture diagram as a clean SVG with proper node positioning was also tricky.

Accomplishments that we're proud of

A fully working AI tool built in under 3 hours, completely free, with 11 features including live architecture diagrams and an interactive installation wizard.

What we learned

How to chain multiple AI API calls efficiently and how to parse and render dynamic AI-generated content in React.

What's next for RepoGuide AI

PDF export, dark/light mode toggle, and a shareable link for every generated guide.

Built With

• chart.js
• github-rest-api
• groq
• llama-3.3-70b
• react
• tailwind-css
• vite