skills.rest
tylern91
tylern91COMMUNITY
Explore all skills by this author →
WebsiteGitHub

k8s-debug

Diagnose Kubernetes workload failures and cluster health issues with deterministic, safety-first workflows.

Updated Aug 23, 2026
Search Tags:#workflow#debugging#security#kubernetes#diagnostics#cluster#pod
One-click install
npx skills add https://github.com/tylern91/dotfiles --skill k8s-debug-tylern91
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: k8s-debug
Source: https://github.com/tylern91/dotfiles/tree/main/claude/.claude/skills/k8s-debug
Command: npx skills add https://github.com/tylern91/dotfiles --skill k8s-debug-tylern91

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

System administrators and engineers often face opaque Kubernetes issues that block deployments and degrade reliability. This Kubernetes Debugging Skill provides a deterministic, safety-first workflow to collect diagnostics, identify root causes, and guide remediation.

Core Features & Use Cases

  • Deterministic debugging workflow that guides you through preflight checks, scope, and targeted diagnostics using built-in scripts (cluster_health.sh, network_debug.sh, and pod_diagnostics.py).
  • Reference-driven troubleshooting with navigable paths to troubleshooting_workflow.md and common_issues.md for symptom-to-fix mapping.
  • Safety-first operations with explicit pre-action snapshots and rollback considerations before disruptive commands.

Quick Start

Execute the Kubernetes debugging workflow by running the included scripts against a problem pod and follow the guided steps to collect diagnostics.

Frequently Asked Questions about k8s-debug

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug Kubernetes pods stuck in CrashLoopBackOff or OOMKilled states?▼

To debug Kubernetes pods stuck in CrashLoopBackOff or OOMKilled, run the pod_diagnostics.py script to collect lifecycle diagnostics, then follow the structured troubleshooting workflow to identify root causes and apply targeted fixes safely.

What is the best way to troubleshoot Kubernetes service connectivity and DNS issues?▼

Troubleshooting Kubernetes service connectivity and DNS issues involves executing the network_debug.sh script to map network failures, followed by referencing common_issues.md for symptom-to-fix mapping and guided remediation.

How do I diagnose Kubernetes cluster health issues during a deployment rollout?▼

Diagnosing Kubernetes cluster health issues during deployment rollouts requires running cluster_health.sh to perform preflight checks, gathering scoped diagnostics to safely isolate workload failures across the cluster.

Can I safely run diagnostics on a production Kubernetes cluster without causing disruptions?▼

You can safely run diagnostics on a production Kubernetes cluster using the safety-first workflow, which mandates explicit pre-action snapshots and rollback considerations before executing any potentially disruptive diagnostic commands.

Why does my Kubernetes pod stay in Pending or ImagePullBackOff status?▼

A Kubernetes pod stays in Pending or ImagePullBackOff status due to resource constraints or image fetch failures; the deterministic debugging workflow uses pod_diagnostics.py to pinpoint the exact lifecycle blockage and guide safe remediation.

Related Skills

Explore More Software Engineering Skills→
paulanunes85paulanunes85
1

kubectl-cli

Execute kubectl commands to diagnose and fix Kubernetes cluster issues in AKS and ARO environments.

Community
Intermediate
filipemottafilipemotta
1

k8s-troubleshooter

Diagnose and resolve Kubernetes issues across clusters and workloads.

Community
Advanced
cnoe-iocnoe-io
392

cluster-resource-health

Query EKS clusters via kubectl and report pod, node, and resource health.

Official
Intermediate
eveldeveld
10

k8s-debug

Attach ephemeral debugging containers to running Kubernetes pods for live inspection.

Community
Intermediate