kube

name	kube
description	Analyze Kubernetes cluster state via SSH. Usage: /kube <key> <user> <host_ip> <namespace>

Kubernetes cluster analysis

Arguments

• Key: $0 (SSH key name in ~/.ssh/: incognia, faraday, kone, cad)
• User: $1 (SSH user, e.g. ubuntu, root, admin)
• Host IP: $2 (server IP address)
• Namespace: $3 (Kubernetes namespace to analyze)

Instructions

1. Read the full CoT: Load and follow ~/rules/cot/kube.md from line 1 to end
2. Read the rules: Load ~/rules/rulesets/KUBE.md for conventions and commands
3. Connect via SSH: ssh -i ~/.ssh/$0 $1@$2 — check ~/.ssh/config for custom port
4. Cluster health:
   • kubectl get nodes -o wide
   • kubectl top nodes (skip if Metrics API unavailable)
5. Namespace analysis ($3):
   • kubectl get pods -n $3 -o wide
   • kubectl get svc -n $3
   • kubectl get vs -n $3
   • kubectl get events -n $3 --sort-by='.lastTimestamp' | tail -20
6. Istio status:
   • kubectl get gateway -n $3
   • istioctl analyze -n $3 (if available)
   • istioctl proxy-status
7. ArgoCD:
   • kubectl get app -n argocd | grep $3
8. Manifests on disk:
   • ls /home/$1/kubernetes-$3/services/overlays/dev/
   • ls /home/$1/kubernetes-$3/services/base/
9. Health checks:
   • Get Gateway IP: kubectl get svc -n istio-system istio-ingressgateway -o jsonpath='{.status.loadBalancer.ingress[0].ip}'
   • Test endpoints via Gateway
10. Observability:
    • Kiali: port 20001
    • Prometheus: port 9090
    • Grafana: port 3000
11. Report: Summarize findings — nodes, problematic pods, services, VirtualServices, ArgoCD sync, recent events, observability

Examples

• /kube kone ubuntu 192.168.10.70 stab — infra server, stab namespace
• /kube kone ubuntu 30.30.40.90 sinaloa-qa — client server via kone
• /kube cad admin 10.0.1.50 produccion — client server via cad

References

• Detailed CoT: ~/rules/cot/kube.md
• Rules: ~/rules/rulesets/KUBE.md