K3s Addons
You deploy k3s addons by placing manifest files into a watched server directory. The cluster reads these files and runs the new components for your team. Most teams set up these extras to handle network routing, storage, and metrics. To manage them, admins often hand out broad cluster roles to engineers. When roles are too broad, a user can change critical cluster settings by mistake. A small typo can break routing or expose private data across the cluster. How k3s addons work in your cluster
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.
You deploy k3s addons by placing manifest files into a watched server directory. The cluster reads these files and runs the new components for your team. Most teams set up these extras to handle network routing, storage, and metrics. To manage them, admins often hand out broad cluster roles to engineers. When roles are too broad, a user can change critical cluster settings by mistake. A small typo can break routing or expose private data across the cluster.
How k3s addons work in your cluster
K3s ships with a few built-in extras for basic DNS, storage, and ingress. It also watches a special manifest folder on the server node at all times. When an admin drops a file into that folder, the controller spots it. The controller then creates the pods, services, and roles defined in that file. You can also turn off default extras when you first start the server.
Teams often use Helm charts to install more tools through this same folder. The cluster runs a small job that installs or updates the chart for you. To check or fix these apps, engineers need a cluster config file. Admins usually bind user groups to standard role rules inside the cluster. These role rules check if a person can read or write cluster objects. If the role matches, the API server lets the user run the request.
What native cluster roles solve and what they miss
Native role rules do a good job of blocking unknown users at the door. They let you split read access from write access across different namespaces. You can allow a developer to view pod logs while blocking deletes. This setup decides who can log in and which endpoints they can reach.
Still, static role rules leave a big gap during live cluster work. Once a role grants access, the cluster trusts every command inside that scope. Native roles cannot inspect the exact commands an engineer runs inside a pod shell. They also cannot rewrite sensitive fields like emails or card numbers in log outputs. Admins must either give out wide admin keys or approve every tiny change by hand. Static config files also sit on user laptops where keys can leak.
Addons: architecture patterns and best practices
Continue reading? Get the full guide.
Free. No spam. Unsubscribe anytime.
How to tighten access for day-to-day cluster changes
A better approach separates cluster boot tasks from daily debugging work. Let automated pipelines place files into the watched server folder during deploys. No human should need direct host access just to update a cluster chart. Keep host keys locked away and out of daily developer workflows.
For live checks, strip back broad cluster admin bindings from human accounts. Scope each role down to the single namespace the team owns. When someone needs to run a sensitive command, require a quick peer review first. Also make sure you log every command and output at the command level. Clear logs help your team spot mistakes fast and pass security audits with ease.
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
How do you disable built-in k3s addons at startup?
You can pass a disable flag when you start the server process. This stops the controller from loading the default manifest for that specific component. You can then install your own preferred tool for ingress or storage instead.
Where do you put custom manifests for auto deploy?
Place your manifest files inside the server manifest directory on the control node. The built-in controller scans this path and applies any changes it finds. Avoid editing deployed objects by hand because the controller will overwrite your edits.
Why are static kubeconfig files risky for managing cluster tools?
Static files often hold long-lived keys with wide admin rights across the cluster. If a laptop is lost, an attacker can use that file to read secrets. Using short sessions with command checks keeps your cluster much safer.
Related guides
In this post