kubernetes

# Kubernetes ## Overview Use this skill when the user needs declarative deployment guidance for Kubernetes workloads. It is for manifests, rollouts, probes, configuration boundaries, workload decomposition, and day-two concerns like debugging and rollback. This skill should favor small, explicit manifests over magical abstraction layers. Good output should make ownership, selectors, failure modes, and rollout behavior obvious. ## When to Use This Skill - Use when creating or reviewing manifests for Deployments, Jobs, Services, Ingress, or related objects. - Use when the user needs readiness, liveness, scaling, configuration, or rollout behavior clarified. - Use when a deployment works in theory but fails under cluster realities such as probe timing, labels, or config drift. ## Core Concepts ### Labels and Selectors Are the Wiring Many Kubernetes failures are really ownership failures. Stable, semantic labels and selectors matter as much as the pod spec. ### Keep Manifests Minimal and Observable Small manifests are easier to review, diff, and roll back. Add only the configuration needed for the stated workload, then document what validates success. ## Step-by-Step Guide ### 1. Define the Workload Boundary State the container image, entrypoint, ports, config sources, secret sources, and persistence expectations before writing manifests. ### 2. Build the Resource Set Choose the minimum set of resources required. For a typical app this means Deployment, Service, Co...