Lee Ryeong

Korean indie hacker 💪
Gym + Code every day
Obsessed with digging deep into AI security
— one deterministic proof at a time.

Free & Open Source. Test only the apps you own.

🔗https://github.com/ghkfuddl1327-wq
🔗 X: https://x.com/OHS1327

Recent activity

"I stopped using AI to judge AI security. Here's what I do instead."

I tested whether agents leak secrets without being asked — and whether my scanner catches it when they do

I tested whether agents leak secrets without being asked — and whether my scanner catches it when they do

I tested whether agents leak secrets without being asked — and whether my scanner catches it when they do

I tested whether agents leak secrets without being asked — and whether my scanner catches it when they do

I tested whether agents leak secrets without being asked — and whether my scanner catches it when they do

0.2.0: I shipped the coverage my own page had already promised

0.2.0: I shipped the coverage my own page had already promised

A leaked agent key is two debts — and I caught my own scanner lying about it

"audit the detection, not the reasoning"

Your AI agent can refuse to leak a secret — and leak it anyway, in its "thinking"

Your AI agent can refuse to leak a secret — and leak it anyway, in its "thinking"

The leak usually isn't in the answer. It's in the thinking.

Your AI agent can refuse to leak a secret — and leak it anyway, in its "thinking"

Your AI agent can refuse to leak a secret — and leak it anyway, in its "thinking"

Your AI agent can refuse to leak a secret — and leak it anyway, in its "thinking"

Your AI agent's leak risk depends more on the model than the prompt

"I looked into why APIs actually go down. It's rarely a hack — it's an unhandled exception."

I tested whether "just paste the leak into your AI to fix it" actually works. It depends on the model — here's what broke.Build-in-public

My AI agent leaked a secret in a way my own scanner missed. Here's what I learned about what these tools can and can't catch.