k8s-debug

What problem does it solve?

This Skill provides a structured approach for diagnosing Kubernetes pod and deployment issues by prioritizing events over logs to speed up root-cause analysis.

Core Features & Use Cases

• Event-first debugging: check pod events before inspecting logs to identify the root cause of issues like CrashLoopBackOff, ImagePullBackOff, or scheduling failures.
• Stepwise investigation flow: a repeatable sequence using common kubectl commands (get_pod_events, get_pod_logs, describe_deployment) to quickly verify state and progress.
• Problem coverage: handles common patterns including CrashLoopBackOff, OOMKilled, ImagePullBackOff, Pending Pods, Readiness/Liveness probe failures, evicted pods, and rollout issues.
• Operational checklist: verify memory/resource limits, recent changes, configuration/secrets, and external service dependencies.

Quick Start

Start debugging a failing pod by listing its events, then fetch logs only if events don’t explain the issue. Move through the recommended steps: get_pod_events → get_pod_logs → describe_deployment.