Use this for writing Kubernetes manifests (YAML), setting up Deployments, Services, Ingress, ConfigMaps, and managing Helm charts.
View on GitHubskills/15-k8s-orchestrator/SKILL.md
February 4, 2026
Select agents to install to:
npx add-skill https://github.com/k1lgor/virtual-company/blob/main/skills/15-k8s-orchestrator/SKILL.md -a claude-code --skill k8s-orchestratorInstallation paths:
.claude/skills/k8s-orchestrator/# Kubernetes Orchestrator
You define robust Kubernetes resources for scalable applications.
## When to use
- "Write a K8s deployment for..."
- "Create an Ingress rule."
- "Set up ConfigMaps and Secrets."
- "Helm chart template."
## Instructions
1. Manifest Structure:
- Use standard API versions (apps/v1 for Deployments, v1 for Services).
- Include metadata labels and selectors for matching pods and services.
2. Deployments:
- Specify resource requests and limits to prevent resource starvation.
- Use liveness and readiness probes to manage pod lifecycle.
3. Configuration:
- Externalize configuration using ConfigMaps.
- Use Secrets for sensitive data (never in plaintext env vars).
- Refer to them via valueFrom in environment variables.
4. Services & Networking:
- Use ClusterIP for internal communication.
- Use Ingress for HTTP(S) routing to external services.
## Examples
User Request: "Write a Deployment for my Nginx server."
Agent Action/Output:
```yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx-deployment
spec:
replicas: 3
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.21
ports:
- containerPort: 80
resources:
requests:
memory: "64Mi"
cpu: "250m"
limits:
memory: "128Mi"
cpu: "500m"
```
User Request: "Expose the deployment with a Service."
Agent Action/Output:
```yaml
apiVersion: v1
kind: Service
metadata:
name: nginx-service
spec:
selector:
app: nginx
ports:
- protocol: TCP
port: 80
targetPort: 80
type: ClusterIP
```Issues Found: