▌BestMCPTools.org
Kubernetes MCP logo

Kubernetes MCP

Let AI agents inspect and manage Kubernetes clusters

Developer ToolsOpen SourceAll ModelsModerate

Last verified: August 17, 2026

Own Kubernetes MCP? Boost it to the top.✦ Upgrade to Featured

Maintenance status

Active
Last code update: September 14, 2026License: MITChecked September 28, 2026

How to Install Kubernetes MCP

$npx -y mcp-server-kubernetes

Requires Claude Desktop, Cursor, Windsurf, or another MCP-compatible client.

Setup by client

Claude Desktop (claude_desktop_config.json)

{
  "mcpServers": {
    "kubernetes-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-server-kubernetes"
      ]
    }
  }
}

Cursor (~/.cursor/mcp.json)

{
  "mcpServers": {
    "kubernetes-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-server-kubernetes"
      ]
    }
  }
}

VS Code (.vscode/mcp.json)

{
  "servers": {
    "kubernetes-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "mcp-server-kubernetes"
      ]
    }
  }
}

Claude Code (terminal)

claude mcp add kubernetes-mcp -- npx -y mcp-server-kubernetes

Add any API keys or environment variables the server's README lists.

About Kubernetes MCP

A Model Context Protocol server that exposes kubectl-style operations to AI assistants: listing pods, reading logs, describing resources, scaling deployments and diagnosing failing workloads through natural language. It reads your existing kubeconfig, so whatever contexts and RBAC rules you already have apply unchanged. Best treated as a read-and-diagnose tool first; write operations should be gated behind a restricted service account. For the container layer underneath, see the Docker MCP setup, and for cluster-wide observability the Dynatrace MCP setup adds problem records and traces.

Key Features

  • kubernetes
  • devops
  • k8s
  • infrastructure
  • observability

Pricing

Open Source: from Free

Pros & Cons

Pros

  • Acts more like a pair programmer with real tools
  • Tight scoping (one repo, dry-run) limits blast radius
  • Keeps flow inside the editor

Cons

  • Write operations require explicit guardrails
  • Quality of output depends on repo hygiene
  • Larger monorepos may need extra filtering

Best For

  • Reviewing diffs and opening PRs from chat
  • Triaging CI failures
  • Scaffolding new packages or modules
  • Repo-wide refactors with verification

Our Take on Kubernetes MCP

Genuinely useful for triage: "why is this pod crashlooping" now takes one prompt. Community-maintained, so scope the service account tightly before giving it write access.

Frequently Asked Questions

Tags