When pods are stuck in CrashLoopBackOff or your deployment rollout hangs at 2am, this gives you a systematic playbook instead of panic-Googling kubectl commands. It includes three diagnostic scripts: cluster_health.sh for cluster-wide snapshots, pod_diagnostics.py for deep pod analysis with logs and events, and network_debug.sh for DNS and connectivity issues. Everything defaults to read-only until you explicitly confirm destructive actions like force-deletes or rollbacks. The workflow guides are organized by symptom, so you can jump straight to "why is my PVC pending" without reading the whole thing. It assumes you have kubectl access but gracefully degrades if optional tools like jq or metrics-server are missing.
npx skills add https://github.com/akin-ozer/cc-devops-skills --skill k8s-debug