🛡️ 100% In-Browser RAM Execution
2-Way Deterministic Secret Tokenizer & AI Prompt Guard
Safely copy-paste code to ChatGPT, Claude, and Gemini without exposing OpenAI API keys, AWS credentials, DB URIs, or private customer data.
🔒
100% Private
Zero files uploaded. Processed entirely in browser memory.
🚫
Zero Advertisements
No popups, redirects, watermarks, or sign-up gates.
⚡
Dual Split Screen
Run side-by-side with other tools on desktop.
🎯 What Can You Do with SecretSanitizer & AI Prompt Guard?
- ✓ Auto-detect 50+ secret types (OpenAI keys, AWS, JWTs, DB URIs, Stripe, private emails).
- ✓ Swap secrets with deterministic {{SECRET_X}} tokens for safe AI prompt sharing.
- ✓ 1-Click re-hydrate AI responses back to original live working credentials.
- ✓ Zero cloud storage — ephemeral mappings exist solely in active session memory.
📋 How to mask and restore secrets in AI prompts with OmniDeck Pro
1
Paste Raw Code with Secrets
Input code containing API keys or DB passwords into Stage 1.
2
Copy Sanitized Prompt to AI
Click 'Mask Secrets' and paste the clean prompt to ChatGPT/Claude.
3
Re-Hydrate AI Response
Paste AI's output into Stage 2 to restore real keys in working code.
💡 Frequently Asked Questions & Privacy Verification
? Are my API keys stored anywhere?
No. Key maps are stored purely in temporary RAM and destroyed when you close the tab.
? How does 2-way rehydration work?
It preserves an in-memory lookup table that replaces tokens with the original values upon return.