Skip to content
AI Dev Toolkit.
Esc
  • AI Token CounterCount tokens for GPT, Claude, Gemini, DeepSeek, Qwen and more.Tool
  • LLM API Cost CalculatorEstimate per-request, daily and monthly API costs.Tool
  • AI Model ComparisonCompare prices, context windows and features across models.Tool
  • AI Model Pricing PagesSpecs, real costs and cheaper alternatives for popular models.Tool
  • Context Window CheckerSee whether your text fits each model’s context window.Tool
  • Subscription vs API CalculatorFind out whether a chat plan or the API is cheaper for you.Tool
  • GPU / VRAM CalculatorCheck how much VRAM a local model needs and which GPUs fit.Tool
  • Prompt Caching CalculatorEstimate savings from prompt caching.Tool

MCP server · Cloud and DevOps

Kubernetes MCP server

Lets an assistant list, inspect and change Kubernetes and OpenShift resources, read pod logs, exec into pods and manage Helm releases.

Maintainer
The containers project
Licence
Apache-2.0
Runs
Locally (npx)
Sign-in
None
Registry version
0.0.67 (updated 2026-09-18)
Last activity
2026-10-13

Overview

What it does

The Kubernetes MCP Server is a native Go implementation that talks straight to the Kubernetes API server instead of wrapping kubectl or helm, so no other tools are needed. It reads your kubeconfig (or in-cluster configuration), picks up changes automatically and can work with several clusters at once. It ships as a single binary for Linux, macOS and Windows, and also as an npm package, a Python package and a container image.

Tools are organised into toolsets. The default config and core sets cover kubeconfig, generic CRUD on any resource, pods (list, get, delete, logs, top, exec, run), namespaces and events. Optional sets add Helm, Tekton, KubeVirt, Kiali, kcp, NetObserv and observability (Loki, Prometheus, Tempo). A TOML configuration file controls everything: read_only = true removes write tools, toolsets limits what loads, and denied_resources can hide kinds such as Secret. It works with OpenShift as well as upstream Kubernetes.

Tools

What the assistant can do

  • Generic create, update, get, list and delete for any Kubernetes or OpenShift resource
  • Pods: list, get, delete, logs, top, exec and run
  • Namespaces, events and OpenShift projects
  • View and manage the current kubeconfig
  • Helm: install, list and uninstall releases (optional toolset)
  • Tekton, KubeVirt, Kiali, kcp, NetObserv and observability toolsets (optional)

Install

Install Kubernetes MCP server in your app

You need:

  • Node.js (for npx), or the native binary, or uv
  • A kubeconfig with access to a Kubernetes or OpenShift cluster

Run in your project. --scope local keeps the servers private to you in this project; use user for every project.

Terminal
claude mcp add --scope local kubernetes -- npx -y kubernetes-mcp-server@latest
Native Windows (PowerShell)
claude mcp add --scope local kubernetes -- cmd /c npx -y 'kubernetes-mcp-server@latest'

Want it alongside other servers? Open Kubernetes MCP Server in the MCP config generator and add the rest; it writes one file per app. The snippets follow the setup in the server’s README, checked on 2026-10-14.

Before you install

  • It acts with your kubeconfig credentials and can delete resources, exec into pods and install Helm charts.
  • Set read_only = true in a TOML file passed with --config, and use denied_resources to block kinds such as Secret; the README points to a read-only ServiceAccount setup for production.
  • A local MCP server runs with your user account’s permissions. Install it only from the source linked above. See how to choose a safe server.

Why it’s listed

Why we list it

A fast, dependency-free server with fine-grained toolsets and a documented read-only mode, released often.

Related

All MCP serversMCP config generatorGuide: MCP servers in Claude CodeGuide: what is MCP?