Naosuu: Redact sensitive identifiers from AI prompts locally

Hey Product Hunt! 👋 I'm Lea, the creator of Naosuu. Like many of you, I love using LLMs to speed up my daily workflows. But as an attorney, I'm intensely aware of the massive privacy and compliance risks that come with pasting text into AI interfaces. Standard setups frequently ingest sensitive data, credentials, and identifiers for model training. I built Naosuu because I wanted a solution that didn't force a compromise between AI utility and total data security. The core philosophy of Naosuu is simple: What happens on your machine stays on your machine. Unlike other compliance tools that route your text through another third-party API to scrub it, Naosuu uses a local-first architecture. It intercepts your prompts, redacts PII, financial data, and custom identifiers locally in your browser, and gives you complete control over your data posture. I’d love to hear your feedback on our redaction accuracy. I’ll be hanging out in the comments all day to answer technical questions about our local architecture, privacy frameworks, or future roadmap. What custom redaction rules would make your daily AI workflows completely stress-free? Let me know below! 👇