Kubernetes agent skills for Claude Code, Cursor and RevampKey

5 ready-made playbooks that teach an AI agent how to do real Kubernetes work: which tools to call, what to check first, and when to stop and ask you.

Download RevampKey, free I have it: connect Kubernetes

Connect Kubernetes once and all 5 skills install themselves and stay up to date. Mac, Windows and Linux.

All 5 Kubernetes skills

Each skill is one job. You ask in plain words and the agent loads the skill that matches, then follows its steps.

Start hereAsks before writing22 Kubernetes tools

Kubernetes MCP Server essentials

Teaches an AI agent to confirm the cluster context and namespace, pick read tools before write tools, and ask for explicit approval before any change to a live cluster.

Use it when

You are connecting an AI assistant to a cluster for the first time
You want the assistant to know which Kubernetes tools are safe to run without asking
A command failed with a permission or connection error and you need to know what to do next

Tools it uses

kubectl_context
ping
kubectl_get
kubectl_describe
kubectl_logs
explain_resource
list_api_resources
kubectl_generic
kubectl_apply
kubectl_create
kubectl_patch
kubectl_delete
kubectl_scale
kubectl_rollout
install_helm_chart
upgrade_helm_chart
uninstall_helm_chart
node_management
exec_in_pod
port_forward
stop_port_forward
cleanup

Example

Check which cluster I'm pointed at, then tell me what's running in the payments namespace.

You are on context prod-eu-1, cluster prod-eu-1, namespace payments. Listing that namespace (read only, no changes made) shows 12 pods: 10 Running, 2 in CrashLoopBackOff (checkout-worker and refund-job). I have not changed anything. Want me to pull the last logs from checkout-worker?

Use in RevampKeyRead the full skill →

Read-only6 Kubernetes tools

Triage Kubernetes cluster health with AI

Sweeps a cluster or namespace for unhealthy pods, crash loops, pending workloads, failed rollouts and node problems, and ranks them by severity. It is read-only and changes nothing.

Use it when

A deploy went out and something seems wrong, but you do not know what
You want a quick health check of a namespace before a release
Users report slowness and you need to know whether the cluster is the cause

Tools it uses

kubectl_context
kubectl_get
kubectl_describe
kubectl_rollout
node_management
kubectl_generic

Example

Triage the payments namespace on the current context and tell me what is most urgent.

Context prod-east, namespace payments, swept just now. Overall: degraded. Critical: checkout deployment has 0 of 3 replicas ready and has restarted 212 times, last restart 09:14. High: refunds worker is at 1 of 2 ready and crash looping on a config error. Medium: 4 pods Pending on insufficient memory across two nodes. Low: 9 completed pods and 3 evicted pods, counted but not listed. Nothing was changed.

Use in RevampKeyRead the full skill →

Read-only12 Kubernetes tools

Audit Helm releases and deployed workloads in a namespace

Inventories Helm releases in a namespace, compares them with what is running, and flags failed upgrades, drift and stale versions for review. It is read-only and makes no changes.

Use it when

You want to see which Helm releases in a namespace are failing or stuck
You suspect someone changed workloads by hand and want to find the drift
You need a list of unowned or outdated workloads before a cluster review

Tools it uses

kubectl_context
install_helm_chart
kubectl_get
kubectl_describe
kubectl_patch
kubectl_scale
kubectl_logs
upgrade_helm_chart
uninstall_helm_chart
exec_in_pod
port_forward
cleanup

Example

Audit the Helm releases in the payments namespace on the staging cluster and tell me what is drifted or failed.

Staging context confirmed, namespace payments. 6 releases: 5 deployed, 1 stuck in pending-upgrade (billing-api, revision 14, still serving revision 13). Most urgent: billing-api's upgrade stalled on an image pull error, so a rollback or retry decision is needed. The reporting-cron deployment runs a tag two versions behind the chart default, so check upstream. No changes made.

Use in RevampKeyRead the full skill →

Asks before writing17 Kubernetes tools

Safe change rollout for Kubernetes with AI

Plans one scoped change to a workload, such as a scale, patch, manifest apply or rollout, with a snapshot and a rollback path. It asks for your explicit approval before every write.

Use it when

Scaling a deployment up or down while keeping a rollback path
Applying a manifest change after checking it against the live object
Rolling back a bad release to a named revision

Tools it uses

kubectl_context
kubectl_get
kubectl_describe
kubectl_rollout
kubectl_generic
explain_resource
node_management
kubectl_apply
kubectl_patch
kubectl_scale
kubectl_create
kubectl_delete
install_helm_chart
upgrade_helm_chart
uninstall_helm_chart
exec_in_pod
kubectl_logs

Example

Scale the checkout deployment in the prod-eu namespace from 3 to 5 replicas, and show me how to undo it if latency goes up.

Target: prod-eu-cluster / prod-eu / deployment/checkout. Change: replicas 3 -> 5. I captured the current spec and the rollout history (revision 14) as a snapshot. Risk is medium because the extra pods draw on node capacity. Rollback is a scale back to 3. Waiting for your explicit yes to apply this scale.

Use in RevampKeyRead the full skill →

Asks before writing12 Kubernetes tools

Troubleshoot a failing Kubernetes workload with AI

Diagnoses one crashing, pending or unready pod, deployment or job, from state and events through logs, and proposes a fix. It never applies a change without your approval.

Use it when

A pod is stuck in CrashLoopBackOff and you need the cause, not just the status
A deployment has replicas that will not become Ready after a rollout
A job keeps failing and you want to know whether the image, config or memory is at fault

Tools it uses

kubectl_context
kubectl_get
kubectl_describe
kubectl_logs
exec_in_pod
kubectl_apply
kubectl_patch
kubectl_delete
kubectl_scale
kubectl_rollout
kubectl_create
kubectl_generic

Example

Why is the checkout-api deployment in the prod namespace stuck in CrashLoopBackOff? Find the cause and suggest a fix.

Checkout-api has 1 of 4 pods crash looping, restarting 27 times since the 14:20 rollout. The previous container exited with code 137 and the events show OOMKilled against a 256Mi memory limit. Proposed fix: raise the memory limit to 512Mi. I have not applied it; that needs your explicit approval.

Use in RevampKeyRead the full skill →

Get all 5 without copying a file

RevampKey adds the skills when you connect Kubernetes, updates them for you and removes them when you disconnect.

Download RevampKey

How Kubernetes skills work

Connect Kubernetes
The Kubernetes MCP server gives the agent its tools.
Ask in plain words
“Triage the payments namespace on the current context and tell me what is most urgent.” Each SKILL.md says when it applies, so only the one that fits is loaded.
Approve any change
The agent calls the tools in order, checks its work, and shows you any change before it makes it.

How to install Kubernetes skills

In RevampKey the skills install themselves when you connect Kubernetes. In other tools each skill is one folder with a SKILL.md file, and you also need the Kubernetes MCP server connected there.

About 1 minute · nothing to copy

Download RevampKey for Mac, Windows or Linux and sign in.
Open Connectors and connect Kubernetes. All 5 skills come with it and stay up to date; they go away when you remove the connector.
Ask in plain language, for example “Triage the payments namespace on the current context and tell me what is most urgent.”
Download RevampKeyConnect Kubernetes · 1 click
Get RevampKey to connect Kubernetes
RevampKey is a free desktop app for Mac, Windows and Linux. Install it, sign in, and Kubernetes is one click away with its skills already attached.
Download RevampKey
Not now? Use the skills in your current tool
for s in essentials cluster-health-triage release-and-helm-audit safe-change-rollout workload-troubleshooting; do curl -fsSL --create-dirs -o .claude/skills/kubernetes-mcp-server-$s/SKILL.md https://revampkey.com/marketplace/kubernetes-mcp-server/skills/$s/SKILL.md done
Other tools

Into this project’s .claude/skills folder

Connect the Kubernetes MCP server in Claude Code.

Run this in your project folder to add all 5 skills:

for s in essentials cluster-health-triage release-and-helm-audit safe-change-rollout workload-troubleshooting; do
  curl -fsSL --create-dirs -o .claude/skills/kubernetes-mcp-server-$s/SKILL.md https://revampkey.com/marketplace/kubernetes-mcp-server/skills/$s/SKILL.md
done

Ask in plain language, for example “Triage the payments namespace on the current context and tell me what is most urgent.”

Into the shared .agents/skills folder

Connect the Kubernetes MCP server in Cursor.

Run this in your project folder to add all 5 skills:

for s in essentials cluster-health-triage release-and-helm-audit safe-change-rollout workload-troubleshooting; do
  curl -fsSL --create-dirs -o .agents/skills/kubernetes-mcp-server-$s/SKILL.md https://revampkey.com/marketplace/kubernetes-mcp-server/skills/$s/SKILL.md
done

Ask in plain language, for example “Triage the payments namespace on the current context and tell me what is most urgent.”

Into the shared .agents/skills folder

Connect the Kubernetes MCP server in Codex.

Run this in your project folder to add all 5 skills:

for s in essentials cluster-health-triage release-and-helm-audit safe-change-rollout workload-troubleshooting; do
  curl -fsSL --create-dirs -o .agents/skills/kubernetes-mcp-server-$s/SKILL.md https://revampkey.com/marketplace/kubernetes-mcp-server/skills/$s/SKILL.md
done

Ask in plain language, for example “Triage the payments namespace on the current context and tell me what is most urgent.”

Into the shared .agents/skills folder

Connect the Kubernetes MCP server in VS Code.

Run this in your project folder to add all 5 skills:

for s in essentials cluster-health-triage release-and-helm-audit safe-change-rollout workload-troubleshooting; do
  curl -fsSL --create-dirs -o .agents/skills/kubernetes-mcp-server-$s/SKILL.md https://revampkey.com/marketplace/kubernetes-mcp-server/skills/$s/SKILL.md
done

Ask in plain language, for example “Triage the payments namespace on the current context and tell me what is most urgent.”

RevampKey or copy the files?

In RevampKey

One click when you connect Kubernetes
Updated for you when a skill improves
Removed when you disconnect
The MCP server is set up for you
800+ other connectors, same way

Download free

Copy into your project

One command per project
Run it again to get updates
You connect the Kubernetes MCP server yourself
Works in Claude Code, Cursor, Codex and VS Code

See the command

Kubernetes skills FAQ

What is a Kubernetes agent skill?

A SKILL.md file: a short playbook an AI agent loads when your request matches it. Each Kubernetes skill describes one job, which Kubernetes tools to use in which order, what to check first and when to stop and ask you.

Do the Kubernetes skills change anything on their own?

2 of the 5 only read. The others tell the agent to show you the change and wait for a yes before it creates, edits or deletes anything in Kubernetes.

Which tools can use these skills?

Any agent that reads the open Agent Skills format: Claude Code, Cursor, Codex, VS Code with Copilot and RevampKey. In RevampKey they are attached when you connect Kubernetes; elsewhere you copy the folders into your project or skills directory.

Do I need the Kubernetes MCP server too?

Yes. The skills tell the agent how to do the work; the Kubernetes MCP server gives it the tools to do it. The Kubernetes MCP setup guide covers every client.

Kubernetes MCP server setup · This page as Markdown

Put Kubernetes’s best workflows in your agent

Download RevampKey, connect Kubernetes and ask your first question in about a minute.

Download RevampKey Connect Kubernetes · 1 click

All 5 Kubernetes skills

Kubernetes MCP Server essentials

Use it when

Tools it uses

Triage Kubernetes cluster health with AI

Use it when

Tools it uses

Audit Helm releases and deployed workloads in a namespace

Use it when

Tools it uses

Safe change rollout for Kubernetes with AI

Use it when

Tools it uses

Troubleshoot a failing Kubernetes workload with AI

Use it when

Tools it uses

Get all 5 without copying a file

How Kubernetes skills work

How to install Kubernetes skills

RevampKey or copy the files?

In RevampKey

Copy into your project

Kubernetes skills FAQ

Put Kubernetes’s best workflows in your agent

Skills for related tools