Security
Security
How Infraova protects infrastructure data and operations — encryption, isolation, access control, and audit logging.
Security principles
Infraova operates on infrastructure that belongs to your clients. The security architecture reflects that responsibility. Provider credentials are encrypted and isolated. Access is role-controlled. Every operation is logged. The approval boundary ensures no change to client infrastructure happens without explicit human authorization.
Credential storage
API keys and OAuth tokens for connected providers are encrypted at rest using AES-256 encryption. Credentials are stored in isolated credential storage — they are never written to application logs, never included in error reports, and never exposed after the initial save. When you rotate a credential in Infraova, the previous credential is immediately invalidated in storage.
Data isolation
Organization data is isolated at the data layer. One organization cannot access the data of another. All queries are scoped to the authenticated organization. There is no shared data between organizations — clients, domains, issues, proposals, audit trails, and credentials are all organization-scoped.
Encryption in transit
All communication between Infraova and connected providers uses HTTPS with TLS 1.2 or higher. All communication between your browser and Infraova uses HTTPS. Internal service communication is also encrypted in transit. Infraova does not support unencrypted connections to any endpoint.
Access control and authentication
Infraova uses role-based access control as described in the Access & Permissions documentation. Authentication is handled through secure session management. Sessions expire after a period of inactivity. All login events are logged. Failed login attempts are rate-limited.
Audit logging
Every action taken in Infraova — by operators, by automated processes, and by API clients — is logged in the audit trail. Audit logs are immutable. They are retained for the lifetime of the organization. The audit log is the security record as well as the operational record — it shows who accessed what, when, and what they did.
The approval boundary as a security control
The approval boundary is not just an operational feature — it is a security control. It ensures that even if Infraova's automated systems produce an incorrect proposal, no change reaches infrastructure without a human reviewing and authorizing it. The approval boundary limits the blast radius of any automated error to the proposal stage, before any real-world change is made.
Note
Infraova does not make any changes to connected infrastructure without an approved proposal from an authorized operator. This is enforced at the system level, not just by policy.