MCP server · Developer tools

Kubernetes MCP server

Lets your AI look at and (if you allow it) manage your Kubernetes clusters, safely and read-only by default.

Developer tools For it and development For operations Cloud Platforms

This is a helper that connects your AI assistant to Kubernetes, the system many companies use to run their apps on servers. Once it is set up, you can ask your AI things like which pods are running or why a deployment is failing, and it will look it up for you. It is handy for anyone who works with a Kubernetes cluster but does not want to memorize all the commands.

What is an MCP server? The 30-second version

On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI a new skill or a connection to another app or service. This one connects your AI to Kubernetes, so it can look things up in your clusters or make changes there when you ask. You stay in control because the helper starts in a safe read-only mode and only does more if you turn that on.

What this MCP server does

You ask your AI a question about your Kubernetes cluster, like which pods are running or why something crashed. The AI sends that request to this helper program running on your computer. The helper talks to your cluster using your existing kube-config file, the same settings the kubectl command uses. It brings back the answer, a list, or a summary, and shows it to you in the chat. If you have allowed it, it can also make changes like restarting or scaling a deployment, but only after you turn those on.

Flow diagram: you ask your AI “List all pods in the default namespace”, on your own computer the Kubernetes MCP server works with kubernetes cluster, and you get back plain answer in chat.

Click to zoom

What you can do with it

List your clusters, namespaces, pods, services, nodes, and deployments
Read the logs of a specific pod
See recent events in a namespace to spot problems
Restart or scale a deployment (only if you allow it)
Change the container image of a deployment (only if you allow it)
Apply a manifest file to create or update resources (only if you allow it)
Delete a resource or run a command inside a pod (only in admin mode with extra opt-ins)

Try asking your AI

“List all pods in the default namespace and tell me which ones are not running”
“Show me the last 50 log lines from the pod called web-abc123”
“Why did my deployment api-server restart last night? Check the events”
“Scale the frontend deployment to 3 replicas”

What it gives back to you

You get back plain answers in the chat: lists of pods or namespaces, log lines, event summaries, or a short explanation of what the helper found. When it makes a change, it tells you what it did, for example that a deployment was restarted or scaled. If something is blocked by the safety settings, it says so instead of doing it.

Before you start

What you need

A Kubernetes cluster you can reach and a kube-config file (usually at ~/.kube/config)
Node.js installed on your computer so npx can run the helper
An MCP-capable AI client like Claude Desktop, Claude Code, Cursor, Windsurf, or VS Code with Copilot

Good to know

In read-write or admin mode this helper can change or delete things in your live cluster, so start in read-only mode and turn on delete, apply, and exec only when you really need them.

Install it with your AI

Add Kubernetes MCP server to your AI, no technical skills needed

You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.

Sign in to get the install prompt

Members get a ready-made prompt that lets the Claude desktop app check Kubernetes MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,066 MCP servers here for you.