Admin Operations Console

Unify reliability, trust evidence, evaluation CI, partners, growth, billing, pilots, monitoring, audit, and users

Admin Operations Console

The Admin Operations Console is the V5 command center for launch operators. It pulls every shipped operational report into one surface so admins can see release confidence, evidence quality, evaluation gates, partner health, growth loops, billing correctness, enterprise pilots, monitoring, audit, and users without jumping between hidden APIs.

/admin now opens with the Leadership Moat report above the operations grid. That layer packages autonomous governance, marketplace distribution, enterprise evidence, and outcome learning before operators drill into individual reliability, evidence, partner, billing, or pilot panels.

What It Unifies

  • Reliability Control Plane: release checks, endpoint pass rate, migration drift, incidents, and SLO burn from /api/admin/reliability.
  • Trust Evidence Warehouse: public proof events, privacy-safe rollups, target coverage, and evidence gaps from /api/admin/trust-evidence.
  • Evaluation CI: score, search, recommendation, review, fix-forward, and model-router gates from /api/admin/evaluation-ci.
  • Partner Marketplace: app registrations, webhook subscriptions, delivery success, retry policy, and dead-letter status from /api/admin/partners/marketplace.
  • Growth Flywheel: trust-safe acquisition, activation, expansion, enterprise, and guardrail signals from /api/admin/growth-flywheel.
  • Billing Assurance: credit reconciliation, Stripe webhook replay safety, entitlement drift, plan gates, revenue anomalies, and repairable issues from /api/admin/billing-assurance.
  • Enterprise Pilots: pilot candidates, policy packs, stakeholder reports, demo tenant readiness, success metrics, and executive export readiness from /api/admin/enterprise-pilots.
  • Monitoring and Audit: analytics volume, operational alerts, configured alerting channels, audit export, and user/credit controls.

Operator Flow

  1. Open /admin as an ADMIN or SUPER_ADMIN.
  2. Read the unified operations score and attention count before reviewing individual panels.
  3. Use the priority queue to inspect critical or watch-list surfaces first.
  4. Use the API coverage list as the route map for deeper debugging or scripted checks.
  5. Use the lower sections for detailed reliability, trust evidence, enterprise pilots, user lookup, credit adjustment, and audit export.

Status Rules

  • healthy means the surface is green and no launch-blocking action is currently required.
  • watch means the surface is usable but needs follow-up before making stronger public launch claims.
  • critical means an operator should resolve the issue before an unrestricted launch or release claim.

The unified score is the average of the normalized surface scores. A single critical surface escalates the whole command center to critical; otherwise any watch-list surface escalates it to watch.

Release Guardrails

  • New admin APIs should be added to lib/admin/operations-console.ts instead of creating one-off dashboard cards.
  • Every operation surface must have an explicit endpoint, score, status, primary metric, and operator signal.
  • Regression tests should cover the surface list, escalation rules, and launch-blocking behavior.
  • Route protection must still classify admin APIs in lib/security/route-protection.ts.