agentleFS
Aidas-devAidas-dev/k8s-agent-skills/skills/helm/SKILL.md

helm

Skill4 starsChanged 5 months agoMITView on GitHub

Use when working with Helm — route to the correct sub-skill based on what you need: writing Helm charts or production operations (GitOps, release inspection, OCI, hooks). **Manual helm CLI for prod deploys is NOT recommended — prefer GitOps (Flux, ArgoCD).**

11 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If routing is ambiguous:

    line 34
  2. Distinctive:

    No server-side component (Helm 3+ client-only).

    line 53

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Helm — Skill Router

Latest: v4.2.3 (v3.21.3 support until Nov 2026)

⚠️ Production Warning

Do not use helm install / helm upgrade manually or from CI for production. Manual Helm bypasses drift detection, audit trail, Git-as-truth, and rollback-via-revert.

Use GitOps instead
Flux: HelmRelease (helm.toolkit.fluxcd.io/v2)
ArgoCD: Application with Helm source

Helm CLI is for development, chart writing, debugging, and emergency recovery — not day-2 operations.

Which Sub-Skill?

Match the user's request against the trigger keywords below, then call skill(name="<matched-skill>") to load the sub-skill's full content.

Routing Table

Trigger keywordsUser wants to...Skill to load
chart, template, HelmRelease values, _helpers.tpl, values.yaml, schema, write, create, modify, library chart, chart testing, helm unittestWrite or modify a Helm charthelm-chart
install, upgrade, rollback, inspect, release, status, get values, get manifest, history, failed, troubleshoot, debug, OCI, push, pull, hook, GitOps, Flux, ArgoCD, HelmRelease, driftDeploy, inspect, or troubleshoot releaseshelm-ops

Decision Tree

If routing is ambiguous:

  1. User is writing code (templates, YAML, values, helpers) → helm-chart
  2. User is running commands (install, inspect, debug, push) → helm-ops
  3. User is asking about both? → Load helm-chart first (chart must exist before deploying), then helm-ops

When NOT to load this router

  • User question is about a specific product's Helm chart (e.g. Higress, Vault, MariaDB) → load that product's *-helm skill instead (e.g. higress-helm, vault-helm)
  • User needs Terraform for Helm releases → load harbor-terraform or similar
  • User is working with ArgoCD or Flux generally (not Helm-specific) → load their respective skills
  • User just needs a quick CLI reference → router's Quick CLI Reference section below is sufficient

Architecture

Chart → template engine (Go + Sprig) → rendered manifests → kubectl apply (SSA in v4) → release stored in Secrets

No server-side component (Helm 3+ client-only). Releases live as Secrets in the target namespace.

Helm 4 vs 3

FeatureHelm 3Helm 4
Apply modeClient-side patchServer-side apply (SSA) default
--atomic✓Renamed to --rollback-on-failure
--forceDelete+recreate podsRenamed to --force-replace
Post-rendererArbitrary binaryPlugin name required
StatusBug fixes until Jul 2026, security until Nov 2026Current stable

Quick CLI Reference

CommandPurpose
helm install <rel> <chart> -n <ns>Install (dev/test only)
helm upgrade --install <rel> <chart> -n <ns> -f vals.yamlIdempotent install-or-upgrade
helm rollback <rel> [rev] -n <ns>Rollback
helm uninstall <rel> -n <ns>Delete
helm list -n <ns> [-a]List releases
helm history <rel> -n <ns>Revision history
helm status <rel> -n <ns>Release status
helm template <rel> ./c -f vals.yamlRender locally (no cluster)
helm lint ./c -f vals.yamlValidate chart structure
helm get values|manifest|notes|all <rel> -n <ns>Inspect release
helm show values|chart|readme|all <chart>Inspect chart
helm push .tgz oci://r.example.com/repoPush to OCI
helm dep update|build|listManage dependencies
helm test <rel> -n <ns>Run test pods

Values Override

FlagExample
-f, --values-f prod.yaml -f override.yaml (later wins)
--set--set image.tag=v2.0
--set-string--set-string image.tag=v2.0.0-rc1

Prefer -f over --set. Audit-friendly, composable for environments.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.