Every PR is a learning moment. sphinx-ci generates a code comprehension quiz from your Pull Request diff using AI. Comment /sphinx on any PR, and the Sphinx creates questions based on the actual code changes. The developer answers the quiz — pass it, and the merge is unlocked. Why? Too many PRs get merged with a quick "LGTM" and nobody truly understands the code. sphinx-ci turns every PR into an opportunity to learn — not a gate to pass, but a mirror to reflect on what you're shipping.

Firecrawl CLI — The complete web data toolkit for AI agents

The complete web data toolkit for AI agents

Hey Product Hunt! 👋 I built sphinx-ci because I was tired of seeing PRs merged with a quick "Looks good to me" when nobody actually read the code. We've all been there. The idea is simple: when someone comments /sphinx on a Pull Request, an AI generates a quiz based on the actual code changes. The developer answers a few questions about their own diff — what does this function do? What happens if this fails? What's the side effect here? It's not about gatekeeping or catching people out. It's about that moment where you have to articulate what you understand — and you realize you maybe don't understand it as well as you thought. A few things I'm proud of: - It's completely free and open source - Your API key never touches our website (it stays in GitHub secrets) - If our service is down, your PRs are never blocked - Everything is customizable per repo I'd love your feedback — especially from teams that do code review. Does this solve a real problem for you? What would you change? Try it at sphinx-ci.dev or check the code on GitHub : https://github.com/AGuyNextDoor/... — Martin, Skillberg

Firecrawl CLI — The complete web data toolkit for AI agents

The complete web data toolkit for AI agents