What problem does it solve?
Inconsistent logging, missing trace propagation, unreliable health checks, and fragmented metrics make incident detection, debugging, and alerting slow and error-prone. This Skill defines a single, enforceable observability standard to make telemetry reliable and searchable across services.
Core Features & Use Cases
- Structured logging: Mandatory JSON logs with required fields such as timestamp (ISO8601 with ms), level, traceId, module, action, and message to enable centralized parsing and querying.
- Tracing: Enforced traceId propagation via x-trace-id with guidance for storing trace context (AsyncLocalStorage) so all logs within a request share the same trace identifier.
- Health checks: Prescribed /health/live and /health/ready endpoints with rules for dependency checks and response shapes to support load balancers and readiness gating.
- Metrics: Prometheus-ready metrics list including http_requests_total, http_request_duration_seconds, db_query_duration_seconds, active_connections, and error_total for SLOs and alerting.
Quick Start
Apply the observability standard to the service user-service to enforce JSON structured logs with required fields, propagate traceId via x-trace-id, expose /health/live and /health/ready, and export Prometheus metrics.