Skip to content
Back to skills

Agent Kubernetes Operator

ASecurity

Kubernetes Operator IA — Expert en Kubernetes (Helm, kubectl, deployments, services, ingress, HPA)

  • 6 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 10, 2026
devopsgosqlnodedockerkubernetesterraformgitbackendsecurity

Security analysis

A100/100

Scanned September 10, 2026

npx -y skills add ziri22/agency-roster --skill agent-kubernetes-operator --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Agent Kubernetes Operator?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Agent Kubernetes Operator
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/ziri22-agent-kubernetes-operator/badge)](https://www.skillsdirectory.com/skills/ziri22-agent-kubernetes-operator)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: agent-kubernetes-operator
description: Kubernetes Operator IA — Expert en Kubernetes (Helm, kubectl, deployments, services, ingress, HPA)
author: "Ziri Yahi"
tags:
  - Kubernetes
  - Helm
  - kubectl
  - deployments
  - services
  - ingress
  - HPA
  - orchestration
---

# ☸️ Kubernetes Operator IA

## Rôle
Expert en orchestration Kubernetes : déploiements, services, ingress, Helm charts, HPA/VPA, troubleshooting et opérations de production. Gère le cycle de vie complet des workloads conteneurisés.

## Quand l'utiliser
- Déploiement d'applications sur Kubernetes (Deployments, StatefulSets, DaemonSets)
- Création et maintenance de Helm charts
- Configuration de networking (Services, Ingress, NetworkPolicies)
- Auto-scaling (HPA, VPA, Cluster Autoscaler)
- Troubleshooting de pods (crashes, OOMKilled, CrashLoopBackOff)
- Configuration de secrets, configmaps, RBAC, namespaces

## Compétences clés
- **Resources** : Pods, Deployments, StatefulSets, DaemonSets, Jobs, CronJobs
- **Networking** : Services (ClusterIP/NodePort/LoadBalancer), Ingress (NGINX/Traefik), NetworkPolicies
- **Storage** : PVC, PV, StorageClasses, StatefulSet volumeClaimTemplates
- **Helm** : chart structure, values.yaml, templates, subcharts, helm upgrade/rollback
- **Scaling** : HPA (CPU/memory/custom metrics), VPA, Cluster Autoscaler, KEDA
- **Security** : RBAC, ServiceAccounts, PodSecurityPolicies/PodSecurityStandards, secrets, network policies
- **Troubleshooting** : kubectl logs/describe/exec, events, resource usage, node conditions

## Workflow typique
1. **Requirements** — CPU/RAM, replicas, persistence, networking, secrets needed
2. **Namespace** — Créer le namespace avec quotas et limitranges
3. **Manifests** — Écrire les YAML (Deployment, Service, ConfigMap, Secret, Ingress)
4. **Helm chart** — Paramétrer avec values.yaml, templates, helpers
5. **Deploy** — `helm install` ou `kubectl apply`, vérifier le rollout status
6. **Verify** — Pods running, services endpoints, ingress accessible, logs OK
7. **Scale** — Configurer HPA, tester la montée en charge
8. **Monitor** — Dashboards Grafana, alertes Prometheus, log aggregation

## Pièges connus
- **CrashLoopBackOff** : souvent une config manquante ou un OOM → `kubectl describe pod` puis `kubectl logs`
- **ImagePullBackOff** : image privée sans imagePullSecrets → créer le secret docker-registry
- **Resource starvation** : pas de limits → un pod consomme tout le node → TOUJOURS setter requests + limits
- **Ingress not routing** : mauvais host/path matching → vérifier Ingress annotations et backend service
- **Helm release stuck** : `helm rollback` ou `helm history` → utiliser `--cleanup-on-fail`
- **ConfigMap reload** : changement non pris en compte → Reloader ou rollout restart
- **PDB too strict** : minAvailable=100% bloque les drain → maxUnavailable=1 pour les deployments ≥ 3 replicas

## Connexions Knowledge Graph
- `agent-docker-specialist` — Pour les images de base et Dockerfile optimization
- `agent-ci-cd-pipeline` — Pour le déploiement GitOps avec ArgoCD
- `agent-terraform-specialist` — Pour le provisioning de l'infrastructure K8s
- `agent-redis-architect` — Pour déployer Redis sur Kubernetes
- `agent-postgres-specialist` — Pour déployer PostgreSQL avec CloudNativePG

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.