Kubernetes MCP Server

What is Kubernetes MCP Server?

The Kubernetes MCP Server is a Model Context Protocol server that enables natural language interaction with a Kubernetes cluster through an MCP client. It provides comprehensive Kubernetes operations including pods, deployments, services, namespaces, and cluster management.

How to use Kubernetes MCP Server?

Clone the repository, install dependencies with pip install -r requirements.txt, then run the server via MCP Inspector, install it in Claude Desktop, or execute directly. Once connected, issue natural language commands like “Get all pods in the default namespace” or “Scale the backend deployment to 3 replicas.”

Key features of Kubernetes MCP Server

• Get, describe, and filter pods
• Execute commands in pods
• Get pod logs and metrics
• Scale deployments and manage rollouts
• Expose deployments as services
• Port forward to services
• Create and delete namespaces
• Cordon, uncordon, and drain nodes

Use cases of Kubernetes MCP Server

• Query pod status and logs in natural language
• Scale deployments and manage rollout history
• Create and manage namespaces interactively
• Perform cluster maintenance like node draining

FAQ from Kubernetes MCP Server

What are the prerequisites?

Python 3.7+, kubectl installed and configured, access to a Kubernetes cluster, and an MCP client (e.g., Claude Desktop).

How do I install the server?

Clone the repository and run pip install -r requirements.txt to install dependencies.

How do I run the server?

Use python -m mcp.inspector src/k8s_mcp_server.py for development, python -m mcp.install src/k8s_mcp_server.py to install in Claude Desktop, or run python src/k8s_mcp_server.py directly.

What Kubernetes operations are supported?

Pod, deployment, service, namespace, and cluster operations including get, describe, scale, logs, exec, rollout management, and node scheduling.

Do I need additional authentication?

No; the server uses your existing kubectl configuration for cluster access.