I use AI to write software, but I'm not yet ready to "let go" of understanding the code, so I made this tool which helps me focus my attention on the things that matter in a PR.
I was ready to not like this because yet another AI slop tool, but it's actually pretty neat.
Thanks! I did indeed write it with AI¹. It actually evolved from the approach I was using beforehand: asking every time my coding agent to break down the PR into steps and build an interface for me to go through them (because GitHub PR UI is effectively unusable for me). I got tired of burning tokens building the ~same UI over and over, so I thought "why not define a format for the review and build a decent UI just once?".
¹ Though actually the "harness of guidelines" I've been using recently make Claude not produce _complete_ garbage.
Can you run this entirely locally without uploading artifacts to your website?
Looks very neat, but many companies don’t permit pasting code into external websites without approvals.
It's open source, so yes, though there is no "easy setup" like a Docker image or a Cloudflare template yet.
The artifacts are only kept in your local indexeddb though, so they don't get uploaded. Or well, unless you start a collaboration session, but even then they are encrypted before uploading, and the key stays in your browser (in the url hash), so I can't see any data on my end.
But I definitely understand the concern. Would you prefer it in a Docker image? Or maybe wrapped in an Electron app?
Been trying that myself, but my goal is a better overall review surface compared to GitHub
https://pyor.review
The amount of code AI writes today i don't see its possible to read and review that, and understand the code everything line by line. As the time duration in which it writes is like a human could take 1 year which it write in 2 weeks and with that speed one needs to use newer ways to validate the product. Either build a testing and code review systems or use something like bridge-terminal that could help you do that.