I tested my AI bug hunter on 44 real bugs from Express, ESLint, Hexo and others. It caught 32.

A developer built BuildProof, an AI bug-hunting tool that attempts to prove a bug exists in a code change rather than guess. Tested against the BugsJS dataset of real bugs from projects including Express, ESLint, Hexo, Karma, Shields, Bower and hessian.js, it caught 32 of 44 re-introduced bugs with no hints, and reported no findings on 4 open pull requests in lodash, koa and axios. The tool runs code in a locked, network-free container that is deleted after each check, and offers a free mode using a local Ollama model.

AI writes more and more of our code, and code review can't keep up. So I built BuildProof : a tool that looks at a code change and tries to prove a bug, not just guess one. BugsJS is a public dataset of real bugs from real projects Express, ESLint, Hexo, Karma, Shields, Bower, hessian.js . I re-introduced each bug and let the hunter look at the change, with no hints. It also knows when to stay quiet: I ran it on 4 open pull requests in lodash, koa and axios. It found nothing, and it was right not to. It doesn't invent bugs to look busy. Your code runs in a locked container with no network and is deleted when the check ends. There's also a free mode that uses your own local model Ollama . 30-second demo, no project needed: npx buildproof try On your own git project 3 free cloud checks : npx buildproof hunt --base main --cloud I'd really like feedback, especially on the misses. Which kinds of bugs should it catch that it doesn't?