What we do, what we rely on, and what we have not (yet) been audited for. We prefer to be specific rather than impressive.
StrataCore does not currently hold its own SOC 2, ISO 27001 or PCI DSS certification. We have not been through an independent audit.
CSA STAR Level 1 (self-assessment): in preparation. We will link the registry entry here once published. A Level 1 entry is a self-assessment and is not an audit.
OpenSSF Best Practices (agent): not applicable until the agent is published as open source.
StrataCore runs on third-party infrastructure that publishes its own compliance reports. Their certifications describe their services, not StrataCore.
| Provider | Purpose | Assurance (provider-published) |
|---|---|---|
| Supabase | Database, authentication, realtime | Publishes SOC 2 Type 2 and HIPAA-eligible attestations (see supabase.com/security) |
| Paddle | Payments, tax, invoicing (Merchant of Record) | PCI DSS Level 1 service provider; card data never touches StrataCore |
| Cloudflare | Application hosting (Workers), CDN, TLS termination, inbound email routing | Provider SOC 2 Type 2 / ISO 27001 reports available from the provider (cloudflare.com/trust-hub) |
| Resend | Outbound email: account confirmation, password reset, support replies, notifications | Provider-published security and compliance information at resend.com/security; receives recipient addresses and message content |
| Google (Gemini API) | AI analysis (optional, see AI data handling) | Only used when Platform AI is enabled by an admin |
Workloads (previews, tests, builds) execute on your machines through the agent. The StrataCore control plane stores metadata you or your agent report (service names, status, resource usage, configuration you provide, test results, audit logs). Secrets you save are encrypted at rest with a server-held key, so StrataCore can decrypt them to deliver to your agent. This is not a zero-knowledge design.
AI features are optional. Platform AI is off until an organization admin enables it. It uses the Google Gemini API free tier, where Google states that content may be used to improve its products; credentials are redacted from prompts first, but you should not enable it for sensitive code or logs. Organizations can instead bring their own key (Claude, Gemini, OpenAI and others) so prompts go to a provider account under their own agreement.
Email security@stratacore.dev (see security.txt). Please include steps to reproduce and avoid accessing other customers' data or degrading the service. We acknowledge reports within 3 business days and will not pursue good-faith research that follows these rules.