operate

Production operating model

Prepare authentication, execution, storage, model access, observability, and recovery for production use.

Operationsbeginner0.7.0 Stable
Verified 2026-07-13View sourceReport a docs issue

Before serving users, make sure each part of the request path has a responsible team and a practical health check.

Assign operational responsibility

LayerResponsibilityReadiness evidence
Application serverUser authentication, trusted context, public routesAllowed and denied request tests
Runner serviceAuthenticated execution and public contractsHealth, capabilities, command, and model checks
Runtime and storeDurable sessions, events, terminal outcomes, replayKnown-session completion and later inspection
Model accessApproved provider, model, credential, and deploymentQualification and short test run
ObservabilityCorrelation across application and runnerSearchable request, session, and run identifiers
OperationsIncident action, recovery, and escalationRepeated user-path verification

Release a change

  1. Record the versions and intended behavior.
  2. Verify authentication and denial paths.
  3. Run one terminal and one streaming request through the production route.
  4. Confirm that the result, artifacts, and traces are inspectable after completion.
  5. Exercise cancellation or waiting behavior if the application supports it.
  6. Run the relevant Ruhroh evaluation and product quality gates.

Do not declare readiness from a health endpoint alone. A healthy process can still have broken user context, model access, persistence, or terminal handling.