Kubernetes MCP Server
@basebandit
About Kubernetes MCP Server
An MCP Server for Kubernetes
Basic information
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"kai": {
"command": "docker",
"args": [
"pull",
"cyclon/kai:v1.0.0"
]
}
}
}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 Kubernetes MCP Server?
Kubernetes MCP Server (Kai) is a Model Context Protocol server that exposes Kubernetes operations as MCP tools, letting LLM clients like Claude Desktop, Cursor, and Continue manage clusters through natural language.
How to use Kubernetes MCP Server?
Install via go install github.com/basebandit/kai/cmd/kai@latest or use the Docker image cyclon/kai:v1.0.0. By default it uses your current kubectl context. Configure with Claude Desktop, Cursor, or Continue by adding the server path to their MCP settings. Run with flags like -kubeconfig, -transport, -in-cluster, or -tls-cert to customize behavior.
Key features of Kubernetes MCP Server
- Create, list, get, delete, and stream pod logs
- Manage Deployments, Jobs, and CronJobs
- Create, update, and delete Services and Ingresses with TLS
- Manage ConfigMaps, Secrets, and Namespaces
- Apply or delete raw YAML/JSON manifests, including CRDs
- List, get, cordon, uncordon, and drain nodes; cluster health metrics
Use cases of Kubernetes MCP Server
- List and describe pods, deployments, and services in a namespace
- Create a deployment with specified replicas and image
- Retrieve logs from a specific pod
- Delete a service or apply a YAML manifest
- Create a cronjob that runs on a schedule
- Port forward a service to a local port
FAQ from Kubernetes MCP Server
How does the server connect to a Kubernetes cluster?
It uses your current kubectl context by default. You can specify a custom kubeconfig with -kubeconfig and a context with -context.
What transports are supported?
stdio (default), streamable-http, and sse-legacy (deprecated). For remote use, run with -transport=streamable-http and -sse-addr=:8080.
How can I run the server inside a Kubernetes cluster?
Use the -in-cluster flag so it uses the pod's service account credentials. The recommended way is to deploy via kmcp with an MCPServer resource.
How do I configure it with Claude Desktop?
Add the server command (and optional args) to your claude_desktop_config.json under mcpServers, for example: "command": "/path/to/kai".
Can I secure the HTTP transport with TLS?
Yes, supply the -tls-cert and -tls-key options pointing to your certificate and private key files to enable HTTPS.
More Cloud & Infrastructure MCP servers
Run Model Context Protocol (MCP) servers with AWS Lambda
awslabsRun existing Model Context Protocol (MCP) stdio-based servers in AWS Lambda functions
K8s MCP Server
alexei-ledK8s-mcp-server is a Model Context Protocol (MCP) server that enables AI assistants like Claude to securely execute Kubernetes commands. It provides a bridge between language models and essential Kubernetes CLI tools including kubectl, helm, istioctl, and argocd, allowing AI syste
container-use
daggerDevelopment environments for coding agents. Enable multiple agents to work safely and independently with your preferred stack.
MCP Server that interacts with Azure AI Foundry (experimental)
azure-ai-foundryA MCP Server for Azure AI Foundry: it's now moved to cloud, check the new Foundry MCP Server
🐳 docker-mcp
QuantGeekDevA docker MCP Server (modelcontextprotocol)
Comments