Lee Ryeong's activity on Indie Hackers · Page 3

Recent activity

I tested 5 LLMs for prompt-injection leaks. Same code, 0% to 90%.Building in public — solo, local-zero, validate before monetize.

I spent a day red-teaming my own LLM scanner. Here's what it got wrong.

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

Three AI assistants, three vendors, one bug: the confused-deputy pattern that keeps shipping

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

Rojaprove now ships two live targets you can attack before trusting it — here's why that mattered to me

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

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

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

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

"Your user typed nothing malicious. Your AI leaked their data anyway."

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

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