Claude Kubernetes MCP Server

@Blankcut

About Claude Kubernetes MCP Server

A GitOps Model Context Protocol (MCP) Server.

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "kubernetes-mcp-server-blankcut": {
      "command": "docker",
      "args": [
        "build",
        "-t",
        "claude-mcp-server",
        "-f",
        "Dockerfile",
        "."
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Claude Kubernetes MCP Server?

The Claude Kubernetes MCP Server is a Model Context Protocol server built in Go that integrates with ArgoCD, GitLab, Claude AI, and Kubernetes to orchestrate and automate Kubernetes environments. It exposes a REST API for programmatic interaction with these systems and is configured via a config.yaml file and environment variables.

How to use Claude Kubernetes MCP Server?

Clone the repository, export required environment variables for ArgoCD, GitLab, and Claude, update the config.yaml with credentials and server preferences, then run locally with go run ./cmd/server/main.go or build and run with Docker. All API requests must include the X-API-Key header.

Key features of Claude Kubernetes MCP Server

Kubernetes resource listing (namespaces, pods, deployments, etc.)
ArgoCD application management via API
Claude AI–powered resource analysis and troubleshooting
GitLab commit analysis endpoint
Helm chart for production Kubernetes deployment
Configurable via environment variables and config.yaml

Use cases of Claude Kubernetes MCP Server

Automate Kubernetes troubleshooting by describing issues in natural language
Orchestrate deployments and rollbacks with ArgoCD through Claude
Analyze commits in GitLab for potential impacts on cluster state
Build custom AI-driven chatops for Kubernetes clusters

FAQ from Claude Kubernetes MCP Server

What are the prerequisites to run the server?

Go 1.20+, Docker, a Kubernetes cluster with a valid ~/.kube/config, an EKS cluster with AWS_PROFILE set, ArgoCD credentials, a GitLab personal access token, a Claude API key from Anthropic, and optionally Vault credentials.

How do I authenticate requests to the server?

All API requests must include the X-API-Key header. The API key is set in the config.yaml under server.auth.apiKey and can be interpolated from an environment variable.

What endpoints are available?

The server exposes health check, Kubernetes namespace/resource/event listing, ArgoCD application listing, and several Claude MCP endpoints: resource analysis (/mcp/resource), troubleshooting (/mcp/troubleshoot), commit analysis (/mcp/commit), and a generic MCP request (/mcp).

How can I deploy the server in production?

A Helm chart is included in deployments/helm. Update values.yaml with appropriate secrets and config maps, then run helm install claude-mcp .

Where does data processed by the server reside?

The server itself does not store data; it acts as a proxy between Claude AI and external systems (Kubernetes API, ArgoCD, GitLab, Vault). Data may be temporarily used in requests and responses but is not persisted by the server.

Frequently asked questions

What are the prerequisites to run the server?

Go 1.20+, Docker, a Kubernetes cluster with a valid `~/.kube/config`, an EKS cluster with AWS_PROFILE set, ArgoCD credentials, a GitLab personal access token, a Claude API key from Anthropic, and optionally Vault credentials.

How do I authenticate requests to the server?

All API requests must include the `X-API-Key` header. The API key is set in the `config.yaml` under `server.auth.apiKey` and can be interpolated from an environment variable.

What endpoints are available?

The server exposes health check, Kubernetes namespace/resource/event listing, ArgoCD application listing, and several Claude MCP endpoints: resource analysis (`/mcp/resource`), troubleshooting (`/mcp/troubleshoot`), commit analysis (`/mcp/commit`), and a generic MCP request (`/mcp`).