Security designed around clear access boundaries.

• Merchant pages validate the current session on the server before returning restaurant data.
• Merchant and Partner access use separate server-side session validation.
• Session cookies are HTTP-only and use secure transport settings in production.
• Passwords are stored using the project's salted password-hashing implementation.
• Partner setup links expire and are single-use.
• Login and password-setup attempts are subject to server-side rate limiting.
• Administrative access is revalidated on the server before protected operations run.

GuestSence records restaurant ownership on feedback, follow-up, improvement and related operational data. Server-side authorization and database controls work together to keep restaurant data scoped to the appropriate account.

Partner access is separately scoped to the authenticated Partner identity and does not provide access to private Merchant feedback or another Partner's records.

GuestSence is delivered through Vercel and uses Supabase for database and storage services. Production traffic is served over HTTPS by the deployment platform.

The Supabase service-role credential is read only by server-side application code. OpenAI access also runs through server-side application code. Protected credentials are supplied through environment configuration rather than embedded in public page code.

• Security-sensitive comparisons use timing-safe checks where implemented.
• Guest submissions use unique identifiers and scoped checks to reduce duplicate processing.
• Request rate limiting uses one-way hashed request identifiers.
• Public review redirects keep platform, restaurant and feedback context scoped together.
• AI requests run in server-only workflows with bounded inputs and structured output validation.

Enabled AI-assisted workflows run through server-side application code. GuestSence applies explicit input limits and structured output validation, and selected AI analysis flows remove common email, phone, payment-card and URL patterns from guest-written text before the relevant AI processing step.

AI-generated patterns and suggested actions are also checked against the feedback records available to the workflow before they are presented as evidence-linked output.

If you believe you have found a security issue, email support@guestsence.com with a clear description and enough detail to investigate. Please do not access, change or retain another restaurant's information while testing or reporting a concern.

For information about data handling, see the Privacy Policy.