Classify Kubernetes resource ownership as direct Helm, Flux, or Argo CD.
npx skills add https://github.com/confighub/cub-scout --skill observe-helm
Please help me install this Agent Skill. Skill: observe-helm Source: https://github.com/confighub/cub-scout/tree/main/skills/observe-helm Command: npx skills add https://github.com/confighub/cub-scout --skill observe-helm
This skill resolves confusion about what “Helm-managed” really means for a Kubernetes resource by separating direct Helm installs from Helm applied via Flux or rendered by Argo CD, so you can trace each field back to its true source.
metadata.managedFields manager-string co-signals plus Helm label families to classify ownership as direct Helm, Flux helm-controller (HelmRelease), or Argo CD as a Helm renderer.app.kubernetes.io/managed-by: Helm with app.kubernetes.io/instance, and falls back to helm.sh/chart when needed (including legacy charts).app.kubernetes.io/managed-by: Helm appears, and what the next best diagnostic step should be.Use observe-helm to determine whether a specific Deployment in your cluster is owned by direct Helm, Flux helm-controller, or Argo CD rendering a Helm chart.
High-intent search queries and answers about installing and using this skill.
A Kubernetes resource shows app.kubernetes.io/managed-by: Helm because Argo CD and Flux helm-controller render Helm charts, leaving standard Helm labels on live objects. Analyzing metadata.managedFields manager strings disambiguates the true ownership path and chart provenance.
You can determine Helm ownership by analyzing metadata.managedFields manager identities alongside Helm label signals. This classification applies priority ordering—Flux first, then Argo, then direct Helm—to correctly attribute the Deployment source.
Helm release detection uses the app.kubernetes.io/managed-by: Helm label paired with app.kubernetes.io/instance. The detection mechanism also falls back to the helm.sh/chart label when needed, ensuring legacy Helm charts are properly identified.
ManagedFields manager-string mapping traces Helm chart provenance by reading the manager identities recorded in Kubernetes metadata. It co-signals these manager strings with Helm labels to classify whether the resource originated from direct Helm, Flux HelmRelease, or Argo CD rendering.
Yes, you can trace chart provenance by applying a Helm-via-disambiguation matrix that analyzes managedFields manager strings and Helm label families. It separates direct Helm installs from Flux or Argo CD rendering to identify the exact chart source.
confighubClassify Kubernetes resource ownership via ownerReferences and managedFields signals.
david-driscollOrchestrates Kubernetes deployments via Helm charts, GitOps config, validation, and monitoring checks.
confighubDiagnose Flux-managed Kubernetes resources by tracing ownership to Kustomizations or HelmReleases.
christian-deleonCreate and manage Helm charts for Kubernetes deployments.