Hey Product Hunt 👋 I build and work alongside security and ops teams, and I kept running into the same quiet tension: AI is genuinely great at reading a messy log, a stack trace, or a config dump — but the moment you paste one into ChatGPT or Claude, every internal IP, hostname, username, token and secret in it has just left the building. "Don't use AI" isn't a real policy; people do it anyway because it saves an hour. So I built Data Sanitizer to remove the tradeoff. Paste your text and it detects and replaces every sensitive value with a consistent token (USER_01, HOST_03, SECRET_02…) — 100% in your browser. Nothing is uploaded, no account, works offline. The part I'm most proud of: it's reversible. Paste the AI's answer back and every token maps to its real value, using a mapping that never leaves your device — so you keep the context instead of losing it. It's a web app and a Chrome side-panel extension. Free to try. I'd love blunt feedback from SOC / DevOps / IT folks: what's missing, and what would you need to trust it in your workflow?

Data Sanitizer

Paste logs into AI without leaking secrets or PII

Paste logs into AI without leaking secrets or PII