メニュー
Kubernetes workload discipline — probes, resources, RBAC, disruption budgets. CNCF standard, not vendor; cluster provisioning belongs to infra.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Context engineering — memory navigation, textual graph traversal, progressive disclosure, recovery. Load when assembling context for a task or when lost.
Implementation-side security discipline — headers, dependencies, secrets, authz. Load for hardening work (sec-ops domain).
Commit conventions, branching, anchors in messages, and the git safety rules. Load for any git operation beyond status/diff.
Anti-generic-AI design discipline — bans, tells, and the slop test. Load for ANY new UI surface or visual redesign. Inspired by Impeccable (Apache 2.0).
Motion & animation discipline — purpose, performance, reduced-motion. Load when work involves animation, transitions or scroll effects.
Microcopy discipline — buttons, errors, empty/loading/success states, alt text. Voice and tone come from the project's memory, not from this skill.
| name | kubernetes |
| description | Kubernetes workload discipline — probes, resources, RBAC, disruption budgets. CNCF standard, not vendor; cluster provisioning belongs to infra. |
K8s is a CNCF open standard — ubiquitous enough to be discipline, not vendor. Boundary: workload operation lives here (dev-ops); cluster lifecycle/provisioning is infra (platform-architecture, infra-patterns). Managed-flavor specifics from memory-long §stack.
gitops) — kubectl edit/apply by hand in prod = drift incident · rollout strategy explicit (maxSurge/maxUnavailable are release-engineering choices).resilience).Namespaces are the unit of tenancy: quotas + RBAC + network policies per namespace · RBAC least-privilege per service account — no default SA with permissions, no cluster-admin for workloads, ever (see infra-security identity rules) · NetworkPolicies deny-by-default east-west (see network-design) · secrets as mounted refs from a real secret store, not bare base64 manifests in git.
Stateless by default; StatefulSets only for genuinely stateful workloads (and prefer managed data services — running databases on k8s is an ADR, not a habit) · HPA on metrics that mean something (queue depth > CPU for workers) · anti-affinity for replicas of the same service — three replicas on one node is one replica with extra steps.
kubectl diff-style dry-run in CI before any manifest merges · new workload checklist: probes, resources, PDB, SA, network policy, non-root (see containers) · post-deploy: rollout status + RED metrics green before the report closes (see observability).