Hey Product Hunt 👋 I'm Varun, maker of Opviva. I built it because I kept watching the same thing happen. People are shipping incredible apps with AI — Lovable, Bolt, Cursor, Replit — in days, not months. But AI writes code that works, not code that's safe. I'd look at these apps and find Supabase keys sitting in the browser, databases with no access control, endpoints you could increment to read anyone's data. The builders are brilliant — they just can't see the holes, because the app looks perfect. Every security tool I tried handed me a list of "maybes" and left the work to me. So I built the opposite: an agent you just talk to. You tell Opviva what you shipped (or paste a URL), and it scans your live app, reproduces the actual exploit to prove it's real — not a theoretical warning — opens the fix as a pull request you approve in one click, and keeps watching after launch. The whole idea: ship it, then just say "check it." Security you can talk to, for the 99% of builders who don't have a security team. The scan is free, no signup — I'd genuinely love for you to point it at something you built and tell me what it finds. And I'm here all day for feedback on where to take it next.

Opviva

It proves your app's security holes, then fixes them

It proves your app's security holes, then fixes them