ArgoCD Deploy Helm Chart

Published
Length
3 min read

Luis Felipe Dias

All posts

To run an argocd deploy helm chart flow, point an ArgoCD app at a chart source. ArgoCD renders the chart templates into plain Kubernetes manifests and applies them to your cluster. Most teams set up this flow by linking a Git repo or a Helm registry. They define custom value files and let the controller sync changes on a timer. When cluster roles or sync rules are too broad, one bad commit can break live apps. Even worse, engineers often keep permanent admin keys on laptops to fix failed syncs b

Your coding agents run with your access.

Claude Code reaches prod with a developer's credentials, and approval prompts get clicked through. hoop.dev checks every query, masks sensitive data, and holds risky changes for a human.

To run an argocd deploy helm chart flow, point an ArgoCD app at a chart source. ArgoCD renders the chart templates into plain Kubernetes manifests and applies them to your cluster. Most teams set up this flow by linking a Git repo or a Helm registry. They define custom value files and let the controller sync changes on a timer. When cluster roles or sync rules are too broad, one bad commit can break live apps. Even worse, engineers often keep permanent admin keys on laptops to fix failed syncs by hand.

How an argocd deploy helm chart workflow operates

ArgoCD treats a Helm chart as a template engine rather than a release manager. When you trigger a sync, the repo server fetches the chart and your custom values. It runs a template step to build standard Kubernetes manifests in memory. Next, the controller compares those rendered manifests against the live state of your target cluster. If the live cluster differs from the rendered files, ArgoCD marks the app as out of sync.

Teams can apply those changes by hand or turn on auto sync. With auto sync active, the controller sends the rendered manifests straight to the Kubernetes API. You can also turn on self-healing so the controller overwrites manual cluster edits. Notice that ArgoCD does not use the standard Helm release secrets inside the cluster by default. Instead, it tracks resource state through its own labels and annotations on each Kubernetes object.

What native GitOps solves and what it leaves open

This native workflow solves version drift across your clusters. Every change to a chart version or value file goes through a Git pull request. Reviewers can check the exact changes before they merge the code. The controller also restricts who can trigger a sync through its own role rules. Teams can lock down specific projects so developers only sync apps they own.

Still, a pure GitOps setup only governs changes that enter through Git. It does not control what happens when a chart fails to deploy and an engineer steps in. During an outage, engineers need direct Kubernetes API access to read logs or delete stuck pods. Native cluster roles decide who can log in to the cluster. They do not inspect individual commands or redact sensitive data like emails or card numbers in flight. Giving developers broad cluster keys for debugging creates a big security gap outside your Git pipeline.

Helm Chart Security + ArgoCD Security: architecture patterns and best practices

Continue reading? Get the full guide.

Free. No spam. Unsubscribe anytime.

Managing values and secrets across multiple clusters

Running the same chart across test and live clusters requires clean value separation. Teams store a base values file in Git and layer cluster-specific files on top. When the repo server builds the chart, it merges these files in the exact order you list. This setup keeps your chart dry while letting you tune replica counts per cluster.

Secrets need special care because you cannot store raw keys in plain Git files. Since ArgoCD renders charts without a native secret store, teams pair it with external secret controllers. The chart deploys a custom reference object, and an in-cluster controller fetches the real secret later. Still, any engineer with broad cluster permissions can read that live secret once it lands in Kubernetes.

Govern kubectl and API calls in flight

Identity providers and Kubernetes RBAC decide who logs into a cluster. They cannot redact keys inside returned JSON objects or hold a risky delete call for review. By design, hoop.dev runs as an open-source proxy, a sidecar for your Kubernetes API, checking requests and responses without replacing RBAC.

First, hoop.dev guardrails refuse banned methods, paths, or headers before they run. Next, hoop.dev data masking rewrites sensitive keys in returned Secret, ConfigMap, and log responses in memory, and writes each decision to the audit trail. Explore the repo on GitHub.

Frequently asked questions

Does ArgoCD run Helm install under the hood?

No, ArgoCD renders the chart templates into raw Kubernetes files first. Then it applies those files using standard Kubernetes API calls.

How do I pass custom values to a Helm chart in ArgoCD?

You can point the app definition to specific value files inside your Git repo. You can also write inline value overrides directly inside the app settings.

Why do engineers still need cluster access if ArgoCD deploys everything?

When a chart sync fails or a pod crashes, Git cannot show live runtime errors. Engineers must run direct cluster checks to read logs, inspect events, and fix stuck resources.

The query runs. The emails stay hidden.