I built an AI code reviewer that caught 29 issues in a 200-line PR

I was tired of being the bottleneck on code reviews. Every PR needed me to manually catch the same things — unvalidated inputs, missing error handling, hardcoded secrets, no logging. So I built VetBot.

It installs as a GitHub App on your repo. Open a PR and it automatically reviews the diff, finds production-readiness issues, suggests actual fix code, and blocks merge on critical problems.

I ran it on a deliberately-flawed 200-line PR and it caught 29 issues — 8 critical, including a hardcoded JWT secret, SQL injection vector, and exposed credentials in logs. The clean PR passed with zero false positives.

Pricing: Starter ($49/mo), Team ($149/mo), Scale ($399/mo). Free 14-day trial, no credit card.

Would love to hear what other IH builders think — what would make this a must-have for your workflow?