How Safe Share Protects Your Data
01
Why sensitive information matters
Support tickets and logs often carry emails, tokens, IPs and customer IDs. Once pasted into an AI tool or third-party service, you lose control over where that data goes.
02
Pattern-based detection
Deterministic rules for emails, phone numbers, API keys, tokens, authorization headers, passwords, cards, connection strings and more run in your browser first.
03
AI-assisted analysis
When enabled, the text is sent to Gemini 2.5 Flash for contextual review — but only after every pattern match has been replaced with a placeholder. The AI sees the masked text, not the obvious secrets. Anything not caught by patterns (e.g. a name or project codename) can still be included in what's sent. Turn AI off in Settings to keep everything in the browser.
04
User-controlled redaction
Nothing is removed automatically. Every finding needs your Accept, Reject or Edit. AI suggests. You decide.
05
Final verification
Before export, the safe copy is scanned again. If anything still looks sensitive, Safe Share tells you instead of claiming it's clean.
06
Data handling
No account needed. Scanned text isn't stored by Safe Share. History keeps only metadata (time, type, counts, risk level) in your browser's local storage, which you can clear anytime.
Safe Share is an assistance tool, not a guarantee that all sensitive information will be detected.