kubectl-mcp-server

🥈Silver

kubectl-mcp-server is a Model Context Protocol server for Kubernetes. It enables operations teams to control Kubernetes infrastructure through natural language conversations. Install via npm or pip. Integrates with Claude LLM.

8651550Updated Today

Intermediate30min to implementautomation

Saves ~210 min per use

Quick InstallView Source

git clone https://github.com/rohitg00/kubectl-mcp-server.git

Works with:

Claude

Overview

About This Skill

https://github.com/rohitg00/kubectl-mcp-server

How to Use

[{"step":"Install the kubectl-mcp-server. Run `npm install -g kubectl-mcp-server` or `pip install kubectl-mcp-server` depending on your environment.","tip":"Ensure you have `kubectl` installed and configured with access to your Kubernetes cluster. Test connectivity with `kubectl cluster-info`."},{"step":"Start the MCP server. Run `kubectl-mcp-server` in your terminal or integrate it with your LLM client (e.g., Claude).","tip":"For Claude, add the server to your MCP configuration file (e.g., `~/.claude/settings.json`). Example:\n```json\n{\n \"mcpServers\": {\n \"kubectl-mcp-server\": {\n \"command\": \"kubectl-mcp-server\"\n }\n }\n}\n```"},{"step":"Use natural language to issue Kubernetes commands. For example, ask: 'List all pods in the staging namespace with their status.'","tip":"Be specific with namespaces, resource names, and actions. Include details like `--namespace`, `--label`, or `--output` if needed."},{"step":"Verify the output and take follow-up actions. Review logs, check resource usage, or adjust configurations based on the server's response.","tip":"Use the provided verification commands (e.g., `kubectl get pods`) to cross-check the server's output. If errors occur, check cluster permissions or resource constraints."},{"step":"Automate repetitive tasks. For example, create a script to scale deployments during peak hours or troubleshoot common issues.","tip":"Save frequently used commands as templates in your LLM client or as shell scripts for quick reuse."}]

Use Cases

Quickly diagnose why a pod is crashing and receive actionable recommendations.

Deploy a new application with best practices using simple natural language commands.

Identify and optimize resource usage across all pods to reduce costs.

Audit security settings and permissions across Kubernetes namespaces effortlessly.

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/rohitg00/kubectl-mcp-server

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Use the kubectl-mcp-server to [DESCRIBE YOUR GOAL, e.g., 'list all pods in the production namespace', 'scale the deployment frontend-app to 5 replicas', or 'troubleshoot why the backend-service is failing']. Ensure you specify any required parameters like namespace, deployment name, or resource limits.

Example Output

```
**Action:** Scaling the `frontend-app` deployment to 5 replicas in the `production` namespace.

**Response from kubectl-mcp-server:**
```
✅ Successfully scaled `frontend-app` deployment in namespace `production`:
- Old replicas: 3
- New replicas: 5
- Status: Available (5/5)

**Verification:**
```sh
kubectl get deployment frontend-app -n production
NAME           READY   UP-TO-DATE   AVAILABLE   REPLICAS   AGE
frontend-app   5/5     5            5           5          7d
```

**Next Steps:**
1. Monitor pod logs for any errors:
   ```sh
   kubectl logs -l app=frontend-app -n production --tail=50
   ```
2. Check resource usage to ensure capacity:
   ```sh
   kubectl top pods -n production -l app=frontend-app
   ```

**Note:** If you encounter errors, try:
- Verifying the deployment name: `kubectl get deployments -n production`
- Checking cluster access: `kubectl auth can-i scale deployments -n production`
```