Hey community! :)
Over the last few weeks I built LegacyLens — a tool that takes any public GitHub repository and visualizes it as an interactive "Code City" to help quickly spot technical debt hotspots.
What it does
- Folders become districts, files become buildings
- Building height = LOC(lines of code) + complexity
- Color = risk level (based on my own scoring model)
- Click on a building → detailed Inspector with risk breakdown and AI-generated deep insights + refactoring suggestions
It uses AST parsing for accurate complexity metrics (cyclomatic, nesting depth, large functions, imports etc.) and falls back to heuristic if needed.
Live demo: https://legacylens.iryna.online/
Currently Supported: Python, JavaScript, TypeScript, and Vue.
Don’t see your favorite language here? Let me know! If I see that LegacyLens is actually helping you, I’ll be more than happy to dive into the rabbit hole of other languages. Whether it’s Go, Rust, or even C++, I’m ready to start "building" those cities if you’re ready to explore them! 😄
What I want from the community
Since this is my first bigger project, I’d love your honest feedback:
- Does the Code City metaphor actually help you understand the codebase faster?
- How useful is the risk scoring system?
- Are the AI suggestions practical?
- What would you improve or add?
Looking forward to your thoughts, ideas, and brutal honesty 😄
Thanks for reading!
Top comments (0)