PromptGuard

Detect sensitive content and PII in ChatGPT before you send it.

PromptGuard helps you notice sensitive information before you share it with ChatGPT. It scans live prompts and recent ChatGPT history for personal information, secrets, and sensitive topics such as health, finance, legal, identity, and confidential work content. Detection runs locally in your browser. Prompt and conversation text are not sent to the developer's servers. Key features: - Live warning before sending sensitive prompts - PII detection for emails, phone numbers, API keys, credit cards, SSNs, IBANs, URLs, dates, addresses, and names - One-click token redaction using safe placeholders such as [EMAIL], [PHONE], and [REDACTED_SECRET] - Optional high-accuracy on-device PII model for names and addresses - On-demand ChatGPT history scan - Local privacy dashboard for reviewed conversations - Allowlist and custom keyword settings PromptGuard is designed for privacy-conscious ChatGPT users who want an extra local safety check before sharing personal or confidential information. Category