Kubernetes Observability and Operations Readiness

When to use

Invoke when making a Kubernetes workload or namespace observable and operable for production — scrape wiring, log shipping, tracing, SLO alerts, and runbook inputs — or when auditing an inherited workload that is a black box in incidents.

Do not use for: base manifest authoring (use k8s-workload-packaging-and-manifest); NetworkPolicy/RBAC (use k8s-network-and-identity-policy); autoscaler tuning that consumes these metrics (use k8s-scaling-and-resilience-topology); image hardening and admission (use k8s-supply-chain-and-image-hardening); choosing the observability backend/substrate (decided upstream in infrastructure-platform.md / architecture/operations); cluster provisioning and control-plane audit enablement (out of Family G — owned by the cloud platform stack and Terraform; this skill wires workload/namespace-level collection).

Inputs

Required:

• A workload manifest set from k8s-workload-packaging-and-manifest (the Service/pod this skill makes observable; it exposes a metrics port and stdout logs).
• Approved architecture/operations decisions on the observability substrate, alerting destinations, and runbook hooks, or explicit confirmation they are intentionally deferred.

Optional: