Have you ever wished there was an IMDb Top 250, but for GitHub projects?
That’s the idea that kicked off a recent weekend project of mine. The result? A functional, lightweight, birds-eye spreadsheet navigator of the top 10,000 GitHub repositories, sorted by stars and searchable by topic — kind of like a modern-day directory of the best open-source projects on the internet. We’re calling it the GitHub Top Repos Navigator, and it’s now live at gh.megabyte.space 🧭.
The Stack: Speedrunning a Fullstack App with AI
Building this app was a breeze thanks to modern tools. I bootstrapped the frontend using Bolt AI from StackBlitz, which practically scaffolded the entire Angular + Ionic application for me. It was almost like magic — just a few prompts and the project came together faster than I expected. You can poke around the frontend code here:
👉 GitHub Awesome on StackBlitz
For the backend, I leaned on Rowy, a Firebase-powered spreadsheet backend that makes managing structured data dead simple. It gives us a collaborative spreadsheet interface that anyone familiar with Google Sheets can instantly pick up, while still being developer-friendly and scalable.
The Data Pipeline: Curated and Automated
The heart of this project is its data, which is generated and maintained by a Replit script running on a cron job:
👉 Repl.it GitHub Awesome List Stars Crawler
Every so often, the script randomly explores a selection of hundreds of well-maintained GitHub Awesome lists — these are community-curated lists of top-tier open-source projects across categories like AI, web dev, databases, and more. To keep the quality high, we only include repos that appear in these trusted Awesome lists.
The crawler aggregates metadata (like stars, descriptions, and links) and syncs it with the Rowy backend, ensuring the spreadsheet always reflects the current GitHub landscape.
A Crude but Useful Search
One thing I’ll admit — the search feature is a bit rough around the edges. It works well enough for casual discovery, but you may need to play around with how you phrase your queries. For instance, searching for " AI " might return better results than simply "AI" due to how it tokenizes and matches terms.
Still, it’s functional and fast — and it can help you stumble onto some gems you might otherwise miss.
Try It Now at gh.megabyte.space
You can check out the live project here:
🌐 https://gh.megabyte.space
We’re leaving it online for the community to explore, and we may revisit and expand the tool if there’s interest or feature requests. Want to see filtering by category, stars, or update date? A better search engine? A GitHub integration? Let us know!
Discover more from Megabyte Labs
Subscribe to get the latest posts sent to your email.
