Security you can build on

API keys

• Generated keys are shown once and stored as SHA-256 hashes. Keep them in your secret manager.
• Revoke compromised keys from the console—the invalidation is immediate.
• Create separate keys per environment or integration for granular control.

Provider credentials

• Stored provider tokens are encrypted at rest. Values are masked in the UI and never returned once saved.
• Rotate credentials by overwriting the existing entry or deleting and recreating it.
• Missing credentials surface clear errors both in the console and API responses.

Authentication & access

• Dashboard access uses secure session cookies. Invite only trusted teammates.
• All API requests require a valid Bearer token. Failed attempts return 401 or 403 with guidance.
• Rate limiting protects the platform from abuse. Contact support if you need higher throughput.

Data retention

• Request payloads, responses, cached data, and token usage are stored for observability.
• Purge logs per project if policies require it—contact support for custom retention.
• Avoid logging sensitive data unnecessarily; use cache fields selectively and redact before sending when required.

Compliance & auditing

• Every request captures timestamps, providers, models, token counts, and cached metadata for audit trails.
• Export logs for external archiving or analytics. Integrations with SIEM tools are available on request.
• Maintain least privilege: issue dedicated API keys for each service or team.

Responsible usage

• Respect provider terms of service. ModelRiver acts as an orchestrator and does not override vendor policies.
• Implement user-level rate limits in your application to complement ModelRiver’s platform-wide safeguards.
• Handle personally identifiable information (PII) in accordance with your regulatory obligations.