Claude Kubernetes MCP Server

A Model Context Protocol (MCP) server for Kubernetes cluster management with ArgoCD and GitLab integration.

This repository contains the Claude Kubernetes MCP (Model Context Protocol) server, built in Go. The server integrates with ArgoCD, GitLab, Claude AI, and Kubernetes to enable advanced control and automation of Kubernetes environments.

See Our Documentation Here

Table of Contents

• Overview
• Prerequisites
• Setup Instructions
• Configuration
• Running Locally
• Building and Running with Docker
• Production Deployment
• API Documentation
• Postman Collection
• License

Overview

This server is designed to orchestrate Kubernetes workloads using Claude AI, GitLab, ArgoCD, and Vault. It exposes a REST API that allows programmatic interaction with these systems, driven by a configured config.yaml and authenticated using an API key.

API Documentation

Below are the primary endpoints exposed by the MCP server. All requests require the X-API-Key header:

General

• Health Check
  • GET /api/v1/health

Kubernetes

• List Namespaces
  • GET /api/v1/namespaces
• List Resources
  • GET /api/v1/resources/{kind}?namespace={ns}
• Get Specific Resource
  • GET /api/v1/resources/{kind}/{name}?namespace={ns}
• Get Events for a Resource
  • GET /api/v1/events?namespace={ns}&resource={kind}&name={name}

ArgoCD

• List Applications
  • GET /api/v1/argocd/applications

Claude MCP Endpoints

• Analyze Resource
  • POST /api/v1/mcp/resource
• Troubleshoot Resource
  • POST /api/v1/mcp/troubleshoot
• Commit Analysis (GitLab)
  • POST /api/v1/mcp/commit
• Generic MCP Request
  • POST /api/v1/mcp

All POST endpoints accept a JSON payload containing fields such as:

{
  "resource": "pod",
  "name": "example-pod",
  "namespace": "default",
  "query": "What’s wrong with this pod?"
}

Postman Collection

A ready-to-use Postman collection will be available soon.

Donation

Please contribute to our coffee fund to help us continue to do great things Buy Me Coffee

License

This project is licensed under the MIT License.

Contributing

Documentation will be expanded soon. If you’d like to contribute, feel free to open a pull request or file an issue!

Prerequisites

• Go 1.20+
• Docker
• Kubernetes cluster & valid ~/.kube/config
• EKS cluster with AWS_PROFILE set locally
• ArgoCD credentials
• GitLab personal access token
• Claude API key (Anthropic)
• Vault credentials (optional, depending on use)

Setup Instructions

1. Clone the Repository

git clone https://github.com/blankcut/kubernetes-mcp-server.git
cd kubernetes-mcp-server

2. Export Required Environment Variables

Export credentials for ArgoCD, GitLab, and Claude:

export ARGOCD_USERNAME="argocd-username"
export ARGOCD_PASSWORD="argocd-password"
export GITLAB_TOKEN="gitlab-token"
export CLAUDE_API_KEY="claude-api-key"
export VAULT_TOKEN="optional-if-using-vault"

Ensure a kubeconfig is available:

export KUBECONFIG=~/.kube/config

3. Configure config.yaml

Update kubernetes-claude-mcp/config.yaml with credentials and server preferences:

server:
  address: ":8080"
  readTimeout: 30
  writeTimeout: 60
  auth:
    apiKey: ""${API_KEY}"" 

kubernetes:
  kubeconfig: ""
  inCluster: false
  defaultContext: ""
  defaultNamespace: "default"

argocd:
  url: "http://example.argocd.com"
  authToken: ""
  username: "${ARGOCD_USERNAME}"
  password: "${ARGOCD_PASSWORD}"
  insecure: true

gitlab:
  url: "https://gitlab.com"
  authToken: "${AUTH_TOKEN}"
  apiVersion: "v4"
  projectPath: ""${PROJECT_PATH}""

claude:
  apiKey: "${API_KEY}"
  baseURL: "https://api.anthropic.com"
  modelID: "claude-sonnet-4.5-20250514"
  maxTokens: 8192
  temperature: 0.3

You can use the provided Go templates or environment variable interpolation method.

4. Add API Key for Postman

Please ensure a config.yaml includes an apiKey. This will be used to authenticate requests in Postman or any external client.

Running Locally

cd kubernetes-claude-mcp
go run ./cmd/server/main.go

With Debug Logging:

LOG_LEVEL=debug go run ./cmd/server/main.go --config config.yaml

Server will start and bind to the configured port in config.yaml (default: 8080).

Building and Running with Docker

1. Build the Image

cd kubernetes-claude-mcp
docker build -t claude-mcp-server -f Dockerfile .

2. Run the Container (second build option included)

cd kubernetes-claude-mcp
docker-compose build
docker-compose up -d

Production Deployment

A Helm chart is included in the repository for Kubernetes deployment:

1. Navigate to the Helm Chart Directory

cd kubernetes-claude-mcp/deployments/helm

2. Deploy with Helm

Update values.yaml with appropriate values and run:

helm install claude-mcp .

To upgrade:

helm upgrade claude-mcp .

Please ensure secrets and config maps are properly mounted and secured in the cluster.