GPT-3 Auditor – Scanning code for vulnerabilities with LLMs

Turns out GPT3 is pretty good at finding vulnerabilities (sometimes).

I made this in a few hours at the Scale AI hackathon. So there isn’t a whole lot of syntactic sugar or error handling. For instance, if your code file is too big, I will just skip it. Someone else should maybe figure out how to recursively read long code files or fit it into the context window somehow. That someone is not me, at least not today.

Trial available?No

Leave a Reply

Help us find great AI content


Never miss a thing! Sign up for our AI Hackr newsletter to stay updated.


AI curated tools and resources. Find the best AI tools, reports, research entries, writing assistants, chrome extensions and GPT tools.