Kubernetes

Provides Kubernetes cluster management via MCP with actions for deployments, pods, services, configmaps, secrets, namespaces, metrics, events, and audits.

typescript

GitHub Stars

typescript

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "icy-r-kubemcp": {
      "command": "npx",
      "args": [
        "-y",
        "@icy-r/kube-mcp"
      ],
      "env": {
        "KUBEMCP_CONFIG_SOURCE": "local",
        "KUBEMCP_KUBECONFIG_PATH": "/path/to/kubeconfig",
        "KUBEMCP_RESPONSE_FORMAT": "auto",
        "KUBEMCP_DEFAULT_NAMESPACE": "default"
      }
    }
  }
}

You can manage Kubernetes resources with a dedicated MCP server that lets AI assistants interact with your cluster. It runs locally via a standard MCP client setup and supports default and custom kubeconfig configurations while exposing a rich set of Kubernetes actions.

How to use

To use this MCP server, run it through your MCP client by starting the server package and pointing your client to the generated MCP endpoint. The server supports a default setup that uses your current kubeconfig and a custom setup where you specify a kubeconfig path. You can perform common Kubernetes operations such as listing deployments, pods, services, configmaps, secrets, namespaces, and more. You can also query metrics, events, and audit configurations. When you request destructive actions, you will need explicit confirmation from your client before the action proceeds. You can preview changes with a dry run feature to ensure you understand the impact before applying it.

Steps to use with your MCP client

Install or prepare your MCP client (for example Claude Desktop or Cursor). 2) Configure the MCP server integration by adding entries that run the MCP server locally via a command such as npx with the package name. 3) Start the server from your client and point it to the local or remote MCP endpoint as appropriate. 4) Use the available tools to manage Kubernetes resources through natural language prompts or scripted commands.

What you can do with it

You gain access to a comprehensive set of Kubernetes actions, including deployments, pods, services, configmaps, secrets, namespaces, metrics, events, and audit configurations. You can list resources, retrieve details, create or update configurations, delete resources with safety checks, and observe metrics and events to monitor cluster health.

Available tools

k8s_deployments

list, get, scale, restart, get_status, get_metrics

k8s_pods

list, get, delete, get_logs, summarize_logs, get_status

k8s_services

list, get, get_endpoints

k8s_configmaps

list, get, create, update, delete

k8s_secrets

list, get, create, update, delete

k8s_namespaces

list, get, create, delete

k8s_metrics

get_pod_metrics, get_node_metrics

k8s_events

list, get_resource_events, get_recent_events

k8s_audit

get_config, configure, get_session_log, clear_session_log

Built by

VeilStrat

AI signals for GTM teams