k8s-review

Solid

Review Kubernetes manifests and a specifically verified cluster using bounded, sanitized evidence, then develop evidence-backed optimization hypotheses without changing cluster state. Use for Kubernetes audits, workload health reviews, capacity and scheduling analysis, upgrade-readiness reviews, or optimization assessments; delegate cluster lifecycle and deployment work to k8s-local.

DevOps & Infrastructure 4 stars 1 forks Updated yesterday MIT

Install

View on GitHub

Quality Score: 80/100

Stars 20%
23
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

# Kubernetes Review Review one explicit target without taking ownership of it. A review request never authorizes apply, delete, restart, rollout, scale, patch, drain, cordon, namespace creation, cluster lifecycle, or volume mutation. Compose [k8s-local](../k8s-local/SKILL.md) for lifecycle or deployment work. Use the repository-owned `dot cluster diagnose` collector and its versioned `dot.cluster.diagnostics/v1` schema instead of assembling ad hoc `kubectl` output. Read the [review matrix](references/review-matrix.md) for dimension-specific evidence and exercise the [behavioral evaluations](tests/behavioral-evaluations.md) when changing this workflow. Agent metadata is in [openai.yaml](agents/openai.yaml). ## Target and authority gate 1. Read repository instructions and identify an explicit or repository-derived kubeconfig, context, namespace, and ownership boundary. Never fall back to the user's default current context. 1. Display the intended kubeconfig, context, namespace, and permitted evidence scope. If derivation is ambiguous, stop before any cluster probe. 1. Verify the selected target through `dot cluster --kubeconfig <path> --context <context> status`. Treat a missing tool, absent cluster, unavailable API, context mismatch, namespace mismatch, or authorization failure as a bounded result, not permission to switch targets. 1. For review-only work, keep every operation read-only. If the user later authorizes a mutation, hand lifecycle, deployment, or teardown to `k...

Details

Author
fmind
Repository
fmind/dotfiles
Created
4 months ago
Last Updated
yesterday
Language
Go
License
MIT

Integrates with

Similar Skills

Semantically similar based on skill content — not just same category

DevOps & Infrastructure Listed

k8s

Kubernetes and Helm review and scaffolding for EKS workloads. Use when user says 'review my helm values', 'before I deploy', 'scaffold a new service', 'check values.yaml', or when working in values.yaml, Chart.yaml, or Helm template files.

8 Updated 2 weeks ago
anmolnagpal
DevOps & Infrastructure Listed

kubernetes-specialist

Designs, hardens, and troubleshoots Kubernetes clusters and the workloads running on them. Use when the user says "review this Kubernetes manifest", "why is my pod crash-looping", or "help me design a multi-tenant cluster", or "/agent-collab:kubernetes-specialist." Also offer this proactively when a change touches Deployment/StatefulSet/DaemonSet specs, RBAC bindings, network policies, or persistent volume claims.

0 Updated today
sumitake
DevOps & Infrastructure Listed

kubernetes-troubleshooting

Systematic triage of failing Kubernetes workloads using kubectl. Use when a pod is Pending, CrashLoopBackOff, ImagePullBackOff, OOMKilled, Error, or stuck Terminating; when a PersistentVolumeClaim will not bind; when a Service returns no endpoints or connections are refused; when an app is reachable inside the cluster but not from outside via Ingress or the Gateway API; when a Deployment's replicas never appear because a ResourceQuota, LimitRange, Pod Security, or admission webhook rejected them; when DNS resolution fails inside the cluster; or when a node is NotReady or reporting disk, memory, or PID pressure. Use when a Secret or config is missing or a live change keeps reverting (operator-synced secrets, GitOps reconciliation). Use when someone asks why a workload is not running, not reachable, or not scheduling. Use it too for proactive health checks — when someone asks "is the cluster OK?" or whether something is healthy even though nothing is obviously failing, since a green-looking cluster can still hi

0 Updated 1 months ago
ngaxavi