Code Lantern 🏮
Illuminate Your Code
Inspiration 💡
As developers, we've all faced the daunting task of understanding unfamiliar codebases. Whether joining a new team, reviewing a pull request, or exploring open-source projects, the initial learning curve can be steep. Traditional documentation often becomes outdated, and manually tracing function calls through dozens of files is time-consuming and error-prone.
Code Lantern was born from this frustration. We envisioned a tool that could instantly analyze any codebase and provide:
- Visual architecture maps showing how functions connect
- AI-powered insights about code quality and structure
- Interactive exploration of dependencies and relationships
The name "Lantern" embodies our mission: to shine light into the darkest corners of complex codebases, making the invisible visible.
What It Does ✨
Code Lantern is an automated code analysis platform that transforms raw code into actionable insights:
- Upload & Analyze: Simply upload a
.zipfile of your repository - AI Summary: Receive instant feedback on:
- Architecture strengths and weaknesses
- Technology stack assessment
- Actionable recommendations for improvement
- Interactive Graph: Explore a visual network where:
- Functions are nodes (color-coded by file)
- Function calls are edges (showing dependencies)
- Click any node to trace its execution path with DFS highlighting
- Deep Dive: Browse files and functions with AI-generated descriptions of:
- Input/output parameters
- Purpose and behavior
- Dependencies and call relationships
Key Metrics Dashboard 📊
Code Lantern calculates honest, actionable metrics:
- Health Score: Strict evaluation (0-100) based on complexity and organization
- Complexity Analysis: Average and maximum function complexity
- Language Distribution: Technology stack breakdown
- Project Size Classification: From "Small" to "Enterprise"
How We Built It 🛠️
Tech Stack
Frontend (React + Vite)
- React Router for navigation between Overview, Architecture, and Deep Dive tabs
- Cytoscape.js for interactive graph visualization with compound nodes
- Tailwind CSS for modern, gradient-heavy UI design
- Custom hooks for state management and API integration
Backend (FastAPI + Python)
- FastAPI for high-performance async API endpoints
- Google Gemini 2.0 Flash for AI-powered code analysis
- AST parsing using regex for Python/JavaScript/TypeScript function extraction
- File system management for uploaded repositories
Architecture Highlights
1. Smart Function Extraction
We built custom parsers that handle multiple languages:
# Python function detection
function_pattern = r'^def\s+([a-zA-Z_][a-zA-Z0-9_]*)\s*\('
# JavaScript/TypeScript detection (arrow functions, declarations, etc.)
patterns = [
r'function\s+([a-zA-Z_][a-zA-Z0-9_]*)\s*\(',
r'const\s+([a-zA-Z_][a-zA-Z0-9_]*)\s*=\s*\(',
r'([a-zA-Z_][a-zA-Z0-9_]*)\s*\([^)]*\)\s*=>'
]


Log in or sign up for Devpost to join the conversation.