/kubernetes-health

Kubernetes Health

Dynamic, discovery-driven health diagnostics for Kubernetes clusters

What This Skill Does

• Discovers all installed Kubernetes operators via API discovery
• Dispatches specialized health check agents based on detected APIs
• Aggregates multi-agent results into unified health reports
• Provides weighted health scores (0-100) with status determination
• Generates prioritized recommendations for identified issues

Auto-Trigger Keywords

Primary Keywords

• cluster health
• k8s health
• kubernetes health
• kubernetes diagnostics
• cluster diagnostics
• health check
• health assessment

Secondary Keywords

• node status
• pod health
• deployment health
• operator health
• crossplane health
• argocd health
• cert-manager health
• prometheus health
• api discovery
• cluster status

Error-Based Keywords

• CrashLoopBackOff
• ImagePullBackOff
• Pending pods
• NotReady nodes
• OutOfSync applications
• certificate expiring
• provider not healthy
• managed resource failed

When to Use

Use This Skill When

• Performing routine cluster health assessments
• Troubleshooting cluster-wide issues
• Onboarding to an unfamiliar cluster
• Validating cluster state before/after changes
• Generating health reports for stakeholders

Don't Use This Skill When

• Debugging a single pod (use kubectl directly)
• Modifying cluster resources (this is read-only)
• Real-time monitoring (this is point-in-time)
• Automatic remediation (agents provide recommendations only)

Quick Usage

# Run API discovery
uv run .claude/skills/kubernetes-health/scripts/discover_apis.py

# Get agent dispatch plan
uv run .claude/skills/kubernetes-health/scripts/health_orchestrator.py

# Aggregate results (after agents complete)
uv run .claude/skills/kubernetes-health/scripts/aggregate_report.py

File Structure

kubernetes-health/
├── SKILL.md              # Main skill instructions
├── README.md             # This file (auto-trigger keywords)
├── scripts/
│   ├── discover_apis.py      # API discovery engine
│   ├── health_orchestrator.py # Sub-agent mapping
│   └── aggregate_report.py    # Report aggregation
├── references/
│   ├── operator-checks.md     # Per-operator health checks
│   └── health-scoring.md      # Scoring methodology
└── templates/
    └── health-report.json     # Output schema

Dependencies