Use the Ops dashboard (/ops, requires system.debug.view) for health, readiness, metrics, connector, MCP, and security snapshots.
- backend health and readiness endpoints
- metrics and runtime summaries
- security health snapshot
- connector and MCP visibility
- admin and pilot surfaces that can correlate readiness with adoption and feedback
- richer time-series visualization
- deeper trace and audit correlation in the UI
- clearer evaluation and regression dashboards
- production-grade alerting workflows beyond the current prototype views
During pilot, admins should correlate spikes in errors with audit logs and user feedback (/pilot/metrics and /pilot/readiness).