Why this page
The platform moves fast, and some building blocks are implemented but not yet conclusively verified in every scenario. Instead of scattering such points across many pages, this page lists them in a consolidated, plain-spoken form — so you can make a sound operational decision.
1. Cloud gateways are experimental
Alongside the official vendor APIs there are adapters for the cloud gateways AWS Bedrock, Google Vertex AI and Azure OpenAI. These are implemented but not yet live end-to-end tested.
- Production use of a gateway should happen only after customer-side verification with real cloud credentials.
- The direct vendor APIs (Anthropic, OpenAI, Google Gemini) and on-prem operation via Ollama/vLLM are not affected.
2. Multi-tenancy: foundations in place, full audit-grade separation to come
The tenant foundations are in place: project-boundary isolation, role-based access control (RBAC) and SSO federation (OIDC).
- Full audit-grade multi-tenancy (separate policy, audit, cost and export per tenant) is coming in a later release.
- Until then: one instance per tenant is recommended where hard, auditable separation is required.
3. Execution of generated code — isolated environment recommended
Build runs execute agent-generated code (e.g. mvn verify, npm run build). Such code should always be treated as untrusted.
- Run build runs on an isolated or trusted environment, not on a host with sensitive access.
- The platform offers a container sandbox per run (
execution.sandbox.variant=container, with network, CPU and memory limits). Further sandbox hardening is in progress.
4. Compliance profiles & policy-as-code — enforcement being completed
Policies and compliance profiles are versioned, signed and attested (Ed25519). Provability is therefore in place.
- Hard enforcement of gate strictness and mandatory attestation is being completed in v0.16.
- Until then, attestation and signing are reliable; for enforcing controls also rely on the configured quality-gate modes.