Proposal: Implement CodeRabbit AI for PR reviews across DefiLlama repositories #2270
Description
Proposal: Implement CodeRabbit AI for PR Reviews
Hey,
I’ve been working on a small experiment using CodeRabbit AI - it’s a GitHub app that reviews PRs using LLMs. Since it’s free for open source and pretty lightweight, I tested it on a fork of the DefiLlama app repo to see how well it handles TypeScript/React code.
So far, it’s been decent at catching small issues (like missed dependency arrays, unnecessary re-renders, or unreachable code) and occasionally makes helpful suggestions. It doesn’t replace a human review but could be a useful first pass, especially on community PRs.
You can check out the PR I did with intentional mistakes and coderabbit's review: kanerep#1
No pressure at all, just wanted to share in case there's interest. I’m happy to help set it up if you think it’s worth trying.
Thanks