kubernetes-helm

name	kubernetes-helm
description	This skill should be used when working on Kubernetes or Helm changes in template-ts projects, including values files, env vars, secrets, ingress, resources, hooks, namespaces, pods, and environment-specific deployment settings.

Kubernetes & Helm Workflow

Apply consistent Helm patterns for test and production Kubernetes environments.

Use this skill when

• Update Helm values files
• Add or change environment variables
• Configure resources, probes, ingress, security, jobs, or cronjobs
• Work with Kubernetes secrets, namespaces, pods, and environment URLs

Follow this workflow

1. Locate the target chart in kubernetes/helm/ and edit both values.test.yaml and values.prod.yaml unless the change is environment-specific.
2. Add non-sensitive variables under extraEnvVars with direct value; keep entries alphabetically sorted.
3. Add sensitive variables with valueFrom.secretKeyRef; never inline secrets in values files.
4. Define hook/job secrets through the chart values key hooks.secretName (in values.*.yaml), then consume it in hook templates (for example migration/sync jobs).
5. If the repository uses local env files, mirror new variables in .env.example and .env with safe placeholder defaults.
6. Check chart-specific secret naming and namespace conventions before deploying.
7. Use k8s-tool for runtime checks only after prerequisites are met: k8s-tool is a project wrapper around kubectl with environment shortcuts, and requires an installed CLI plus valid kubeconfig/cluster access.

Reference material

Read concrete snippets and YAML examples in:

• references/chart-structure.md
• references/helm-examples.md

Key Rules

1. Keep extraEnvVars alphabetically sorted
2. Never commit secrets - use K8s secrets with secretKeyRef
3. Test values are conservative - lower resources than prod
4. Use appropriate probe paths - /api/alive for liveness, /api/health for readiness
5. CronJobs need concurrencyPolicy: Forbid to prevent overlapping