Launching today

TestMyPrompt

Launching today

Find vulnerabilities in your AI prompts before your users do

9 followers

Find vulnerabilities in your AI prompts before your users do

9 followers

Visit website

TestMyPrompt helps developers and teams find vulnerabilities in AI prompts before they reach production. Test across 22 security, safety and reliability categories, including prompt injection, jailbreaks, data exfiltration, indirect prompt attacks, tooling permissions, bias and hallucination risk. Get clear results, risk ratings and actionable recommendations to help you ship safer, more reliable AI.

Free Options

Launch tags:Developer Tools•Artificial Intelligence•Security

Launch Team / Built With

Framer AI AgentsDesign and publish professional sites with AI

Promoted

Maker

📌

Hey Product Hunt! 👋 I'm Jae, the developer behind TestMyPrompt. I built it after noticing how much effort we put into testing our code before production, while the prompts powering AI features often don't get the same level of testing. TestMyPrompt is my attempt to make that easier. It currently tests prompts across 22 security, safety and reliability categories — including prompt injection, jailbreaks, data exfiltration, indirect prompt risks, tooling permissions, bias and hallucination risk. It's still early, and that's exactly why I'm launching it here. I'd really value feedback from developers and teams actually building with AI. I'd particularly love to know: → What tests are we missing? → What would make this useful in your development workflow? → Would you want this integrated into CI/CD? → And most importantly... can you break it? 😅 There's a free trial with no card required, so please give it a go. I'll be around all day answering questions and taking feedback. Thanks for checking out TestMyPrompt! 🚀

Report

1d ago

Maker

A quick challenge for anyone checking out TestMyPrompt today 👀

I'd genuinely love to see what you can throw at it.

If you're working with LLMs, take one of your non-sensitive system prompts and run it through TestMyPrompt.

If it:

misses something you think it should catch
flags something you disagree with
gives you a recommendation that isn't useful
or you simply manage to break the tester itself...

please tell me here.

TestMyPrompt is very early, and feedback from people actually building with LLMs is far more useful to me right now than everything working perfectly.

I'll be around throughout the launch and would love to discuss anything you find.

Report

4h ago