Kai
Kai provides a bridge between large language models (LLMs) and your Kubernetes clusters, enabling natural language interaction with Kubernetes resources. The server exposes a comprehensive set of tools for managing clusters, namespaces, pods, deployments, services, and other Kubernetes resources
Kai - Kubernetes MCP Server
A Model Context Protocol (MCP) server for managing Kubernetes clusters through LLM clients like Claude and Ollama.
Overview
Kai provides a bridge between large language models (LLMs) and your Kubernetes clusters, enabling natural language interaction with Kubernetes resources. The server exposes a comprehensive set of tools for managing clusters, namespaces, pods, deployments, services, and other Kubernetes resources.
Features
Core Workloads
- Pods - Create, list, get, delete, and stream logs
- Deployments - Create, list, describe, and update
- Jobs - Batch workload management (create, get, list, delete)
- CronJobs - Scheduled batch workloads (create, get, list, delete)
Networking
- Services - Create, get, list, and delete
- Ingress - HTTP/HTTPS routing, TLS configuration (create, get, list, update, delete)
Configuration
- ConfigMaps - Configuration management (create, get, list, update, delete)
- Secrets - Secret management (create, get, list, update, delete)
- Namespaces - Namespace management (create, get, list, delete)
Cluster Operations
- Context Management - Switch contexts, list contexts, rename, delete
- Nodes - Node monitoring, cordoning, and draining
- Cluster Health - Cluster status and resource metrics
Storage
- Persistent Volumes - PV and PVC management
- Storage Classes - Storage class operations
Security
- RBAC - Roles, RoleBindings, and ServiceAccounts
Utilities
- Port Forwarding - Forward ports to pods and services (start, stop, list sessions)
Advanced
- Custom Resources - CRD and custom resource operations
- Events - Event streaming and filtering
- API Discovery - API resource exploration
Requirements
The server connects to your current kubectl context by default. Ensure you have access to a Kubernetes cluster configured for kubectl (e.g., minikube, Rancher Desktop, kind, EKS, GKE, AKS).
Installation
go install github.com/basebandit/kai/cmd/kai@latest
CLI Options
kai [options]
Options:
-kubeconfig string Path to kubeconfig file (default "~/.kube/config")
-context string Name for the loaded context (default "local")
-transport string Transport mode: stdio (default) or sse
-sse-addr string Address for SSE server (default ":8080")
-log-format string Log format: json (default) or text
-log-level string Log level: debug, info, warn, error (default "info")
-version Show version information
Logs are written to stderr in structured JSON format by default, making them easy to parse:
{"time":"2024-01-15T10:30:00Z","level":"INFO","msg":"kubeconfig loaded","path":"/home/user/.kube/config","context":"local"}
{"time":"2024-01-15T10:30:00Z","level":"INFO","msg":"starting server","transport":"stdio"}
Configuration
Claude Desktop
Edit your Claude Desktop configuration:
# macOS
code ~/Library/Application\ Support/Claude/claude_desktop_config.json
# Linux
code ~/.config/Claude/claude_desktop_config.json
Add the server configuration:
{
"mcpServers": {
"kubernetes": {
"command": "/path/to/kai"
}
}
}
With custom kubeconfig:
{
"mcpServers": {
"kubernetes": {
"command": "/path/to/kai",
"args": ["-kubeconfig", "/path/to/custom/kubeconfig"]
}
}
}
Cursor
Add to your Cursor MCP settings:
{
"mcpServers": {
"kubernetes": {
"command": "/path/to/kai"
}
}
}
Continue
Add to your Continue configuration (~/.continue/config.json):
{
"experimental": {
"modelContextProtocolServers": [
{
"transport": {
"type": "stdio",
"command": "/path/to/kai"
}
}
]
}
}
SSE Mode (Web Clients)
For web-based clients or custom integrations, run in SSE mode:
kai -transport=sse -sse-addr=:8080
Then connect to http://localhost:8080/sse.
Custom Kubeconfig
By default, Kai uses ~/.kube/config. You can specify a different kubeconfig:
kai -kubeconfig=/path/to/custom/kubeconfig -context=my-cluster
Usage Examples
Once configured, you can interact with your cluster using natural language:
- "List all pods in the default namespace"
- "Create a deployment named nginx with 3 replicas using the nginx:latest image"
- "Show me the logs for pod my-app"
- "Delete the service named backend"
- "Create a cronjob that runs every 5 minutes"
- "Create an ingress for my-app with TLS enabled"
- "Port forward service nginx on port 8080:80"
Contributing
Contributions are welcome! Please see our contributing guidelines for more information.
License
This project is licensed under the MIT License.

İlgili Sunucular
Alpha Vantage MCP Server
sponsorAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
Calva Backseat Driver
An MCP server for the Calva VS Code extension, allowing AI assistants to interact with a live Clojure REPL.
Chronos Protocol
A robust MCP server that eliminates temporal blindness in AI coding agents through intelligent time tracking, persistent memory, and complete session traceability.
Blender MCP Senpai
An AI-assisted mentor for Blender that detects topology issues and offers real-time improvement suggestions.
x64dbgMCP
An MCP server that connects LLMs with the x64dbg debugger, enabling natural language control over debugging functions.
Image
Fetch and process images from URLs, local file paths, and numpy arrays, returning them as base64-encoded strings.
n8n-MCP
Provides AI assistants with access to n8n node documentation, properties, and operations.
Inoyu Apache Unomi
Maintains user context and manages profiles using the Apache Unomi Customer Data Platform.
Remote MCP Server (Authless)
An example of a remote MCP server without authentication, deployable on Cloudflare Workers.
ADT MCP Server
An MCP server for ABAP Development Tools (ADT) that exposes various ABAP repository-read tools over a standardized interface.
Universal Infinite Loop MCP Server
A goal-agnostic parallel orchestration framework implementing Infinite Agentic Loop patterns as a Model Context Protocol (MCP) server.