Security at Rhindon Cyber
This page is Rhindon Cyber's own statement of how the Rhindon AI Risk & Integrity Cloud (RAIC) protects customer data. It describes practices that are in place today. Where we make no claim, it is because we are not prepared to assert one.
Tenant isolation
Every organization's data is separated at the database layer. Row-level security policies resolve the signed-in user's organization on each query and refuse rows belonging to any other tenant, and application queries apply a redundant organization filter on top of that so a policy gap alone cannot expose another customer's records. Cross-tenant access exists only for Rhindon Cyber platform administrators acting through an explicit, audited impersonation path.
Access control
Access to the platform requires multi-factor authentication using time-based one-time passwords. Enterprise customers can federate identity through SSO and provision or deprovision accounts automatically with SCIM. Permissions are role-based, and roles are stored in a dedicated table separate from user profiles so a user cannot change their own privileges by editing their profile.
Auditability
Governance and administrative actions are written to an append-only audit log scoped to the organization that performed them, and mirrored to a platform-level log that Rhindon Cyber reviews. Customers can read and export their own organization's audit history at any time; they cannot read another organization's.
Evidence integrity
Governance evidence is recorded in a hash-chained ledger, sealed nightly into a daily root, and signed with ECDSA on the NIST P-256 curve using SHA-256. Anyone holding an exported evidence pack can verify it independently — against the public verification endpoint or entirely offline — without a RAIC account and without trusting Rhindon Cyber. The full scheme is published so the verification is reproducible.
Read the Evidence Verifier Specification or call the public verification endpoint at /api/public/verify-evidence.
Secure development
Every change to the platform runs through automated checks before it can ship.
- An automated test suite covering application logic, access control, and data handling.
- A secret scan that blocks credentials from entering the codebase.
- A dependency vulnerability audit that blocks changes carrying high-severity advisories.
- Periodic review of the codebase against the OWASP Top 10.
Application hardening
Every response the platform serves carries browser security headers that constrain what the page is allowed to do.
| Protection | What it does |
|---|---|
| Content-Security-Policy | Restricts which scripts, styles, fonts, images, and network destinations the app may load, so injected content cannot execute or exfiltrate data. |
| Strict-Transport-Security | Requires HTTPS for the domain and all subdomains for a year, with preload, so a browser never downgrades to an unencrypted connection. |
| X-Content-Type-Options | Stops browsers from guessing a file's type and running it as something it is not. |
| X-Frame-Options | Prevents the app from being embedded in another site's frame, which blocks clickjacking. |
| Referrer-Policy | Limits how much of the current URL is shared when a user follows a link off the platform. |
| Permissions-Policy | Turns off browser capabilities the platform does not use, including camera, microphone, geolocation, payment, and USB access. |
| Cross-Origin-Opener-Policy | Isolates the app's browsing context from windows opened by or opening other origins. |
| Cross-Origin-Resource-Policy | Prevents other sites from loading the platform's resources directly. |
Reporting a vulnerability
If you believe you have found a security issue in RAIC, report it to our security team. Include reproduction steps, the affected URL or endpoint, and what you observed.
Please test only against your own organization. Do not attempt to access another customer's tenant or data, and do not run scans that degrade service for others.
Subprocessors
These third parties process data on RAIC's behalf so the platform can operate.
| Subprocessor | Purpose | Data category |
|---|---|---|
| Managed cloud hosting and database provider | Application hosting, database, file storage, and background jobs | All customer platform data |
| Cloudflare | Edge delivery, TLS termination, DDoS protection, and response-header enforcement | Request metadata and IP addresses in transit |
| Resend | Transactional and notification email delivery | Recipient names and email addresses, message content |
| Stripe | Subscription billing and payment processing | Billing contact details; card data is handled by Stripe and never stored by RAIC |
| Microsoft Azure | Hosts the PII-detection service used by the data-governance scanner | Text submitted for classification, processed transiently and not retained |
| OpenAI | AI model inference for assistant, drafting, and analysis features | Prompt content submitted to those features |
| Google (Gemini) | AI model inference for assistant, drafting, and analysis features | Prompt content submitted to those features |
| hCaptcha | Bot protection on public forms | Request metadata and challenge interactions |
The following integrations are not subprocessors, because the customer directs them into systems the customer already controls:
- Microsoft Graph — tenant discovery scans that the customer authorizes against their own Microsoft tenant.
- ConnectWise PSA — risk events pushed into the partner's own PSA instance.