Kubernetes Manager MCP Server

Overview

The Kubernetes Manager MCP server allows AI assistants to interact with Kubernetes and OpenShift clusters, providing capabilities for resource management, pod operations, namespace/project listing, event monitoring, and Helm chart management.

Installation

Claude Desktop

The easiest way to get started with the Kubernetes Manager in Claude Desktop is using npx. Open your claude_desktop_config.json file and add the following to the mcpServers section:

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

VS Code / VS Code Insiders

You can install the Kubernetes Manager MCP server in VS Code by running:

# For VS Code
code --add-mcp '{"name":"kubernetes","command":"npx","args":["-y","kubernetes-mcp-server@latest"]}'

# For VS Code Insiders
code-insiders --add-mcp '{"name":"kubernetes","command":"npx","args":["-y","kubernetes-mcp-server@latest"]}'

Using npm

If you prefer to install the package globally:

npm install -g kubernetes-mcp-server

Then configure your MCP client to use:

{
  "mcpServers": {
    "kubernetes": {
      "command": "kubernetes-mcp-server"
    }
  }
}

Using Python

The server is also available as a Python package:

pip install kubernetes-mcp-server

Then configure your MCP client to use:

{
  "mcpServers": {
    "kubernetes": {
      "command": "python",
      "args": ["-m", "kubernetes_mcp_server"]
    }
  }
}

Using Docker

You can run the server in a Docker container:

docker run -v ~/.kube:/root/.kube -p 8080:8080 ghcr.io/manusa/kubernetes-mcp-server:latest

Then configure your MCP client to connect to the server at http://localhost:8080.

Configuration

The server automatically detects and uses your Kubernetes configuration from:

• The standard ~/.kube/config file
• In-cluster configuration when running inside a Kubernetes cluster
• Custom configuration specified via environment variables

Environment Variables

• KUBECONFIG: Path to a custom Kubernetes configuration file
• KUBERNETES_MCP_SERVER_PORT: Port to run the server on (default: 8080)
• KUBERNETES_MCP_SERVER_MODE: Server mode, either sse (default) or ws for WebSockets

Usage

Once installed, you can ask your AI assistant to perform various Kubernetes operations. The server provides capabilities for:

1. Configuration Management: View and manage Kubernetes configuration
2. Resource Operations: Perform CRUD operations on any Kubernetes resource
3. Pod Management: List, get, delete, view logs, exec into, and run pods
4. Namespace/Project Management: List Kubernetes namespaces or OpenShift projects
5. Event Monitoring: View Kubernetes events
6. Helm Operations: Install, list, and uninstall Helm charts

Simply ask your AI assistant to perform these operations in natural language, and it will use the Kubernetes Manager MCP server to execute them.