Appinico LLC

On-device PII and brand-term masking for AI chat sites. Prompts are masked before they leave the browser.

NoeticGuard Shield hides personal data and secrets in your messages before they leave your browser for an AI chat. You type or paste an email address, a card number or an API key. Shield replaces it with a token such as [EMAIL_1] on your device. The AI works with the token. When the answer comes back, Shield shows you the original value again, locally. Raw prompts are not uploaded to NoeticGuard for masking. What Shield does • Masks emails, phone numbers, card numbers, national and tax IDs and similar patterns, including US, EU, UK and Turkish ID formats. • Uses the same masking engine as the NoeticGuard SDK, running inside the page. • Follows the policy your workspace sets in the dashboard: which categories to mask and which packs to turn off. • Supports the add-ons your workspace has: Secrets Guard (API keys and tokens become [SECRET_n]), Brand Guard, and the Legal Privilege and COPPA / Student Data dictionaries. • Media Guard (Startup plan and above) blocks image and file uploads and image paste on supported chats. It only blocks. It does not read images and does not run OCR. • Sends usage counts to your dashboard for quota and audit views. Only counts and entity categories. No prompt text, cookies or secrets. • Lets your team report a missed value without pasting the raw value into a ticket. Where it runs Major AI chat sites. The current list and setup notes: https://noeticguard.com/docs/browser-shield Limits • Detection uses rules and can miss values. Treat Shield as one technical control, not a guarantee. • It helps mask the kind of personal data that GDPR, KVKK and CCPA programs care about. This is orientation, not certification. • It does not run in Incognito windows until you allow it there in Chrome's extension settings. • It is a Chrome extension. Desktop apps need a separate IT setup, described in our docs. • It does not rewrite prompts on our servers. If you need that, Shield is the wrong tool. Getting started 1. Create a workspace at https://app.noeticguard.com. The Developer plan is free. 2. Under API Keys, create a publishable key (ng_pub_…). 3. Open the Shield popup, paste the key and check that it says "Connected". 4. Open a supported AI chat and send a test value, such as a made-up email. The model gets a token like [EMAIL_1] instead of the original. Questions Do I need a paid plan? You can start free. Secrets Guard, Brand Guard, Media Guard and the industry dictionaries need a paid plan or an add-on. See https://noeticguard.com/#pricing Will my team still understand the chat? Yes. The same value always gets the same token, and you see the original values restored on your own device. Links Docs: https://noeticguard.com/docs/browser-shield Guides: https://noeticguard.com/guides Privacy: https://noeticguard.com/privacy Security: https://noeticguard.com/security Support: support@noeticguard.com