Media
3 items
Description
Privacy-first guard for sensitive prompt submissions on protected domains. PromptShield helps organizations prevent sensitive data from being pasted or uploaded into protected AI destinations. It evaluates prompt text and file metadata locally in the browser before submission and can warn, redact, require justification, or block activity based on policy. Security teams can centrally manage policy packs, review intervention events, and roll out protected browser sessions. PromptShield is designed for privacy-first operation with browser-side classification, metadata-only event logging by default, and no raw prompt storage in normal event records.