kubeview-debug

What problem does it solve?

This Skill provides structured playbooks to triage Kubernetes clusters step-by-step with KubeView MCP whenever pods crash, rollouts stall, services lose endpoints, nodes report pressure, or resource bottlenecks emerge, all without any write access.

Core Features & Use Cases

• Cluster Triage: Start with kube_list plus kube_metrics to capture overall health, surface namespaces with CrashLoopBackOff or Pending workloads, and highlight high-resource consumers.
• Specialized Investigations: Follow the decision tree to invoke kube_get, kube_logs, kube_log, kube_net, kube_exec, or kube_port for focused diagnostics (pods, deployments, services, nodes, resource usage).
• Advanced tooling: When more logic is needed, leverage plan_step for multi-step coordination or run_code in code-mode to bulk analyze CRDs like pods missing limits while preserving sensitive data.

Quick Start

Use the kubeview-debug skill to triage cluster health by running kube_list diagnostics, gathering kube_metrics summaries, and drilling into kube_get/kube_logs calls for the affected namespaces.