Get detailed information about a specific resource in a namespace
List detailed information about all resources in a namespace
Create or update a resource in a namespace
Delete a resource in a namespace
Describe detailed information about a resource in a namespace
Scale a deployment in a namespace
Execute a command in a pod in a namespace
Get logs from a pod in a namespace
Diagnose all pods in a namespace
Diagnose all deployments in a namespace
Diagnose all statefulsets in a namespace
Diagnose all services in a namespace
Diagnose all cronjobs in a namespace
Diagnose all ingresses in a namespace
Diagnose all networkpolicies in a namespace
Diagnose all validatingwebhooks
Diagnose all mutatingwebhooks
Diagnose all nodes in cluster
Get pod/deployment/replicaset/statefulset resource usage in a namespace (cpu, memory)
Kubernetes Eye provides a powerful interface for managing and diagnosing Kubernetes clusters through an MCP server. It offers comprehensive capabilities for resource management, workload diagnostics, and monitoring, allowing users to efficiently troubleshoot and manage their Kubernetes environments. With support for all native Kubernetes resources and custom resource definitions, this tool enables operations like listing, creating, updating, and deleting resources. It also includes specialized diagnostic capabilities for analyzing the health and configuration of pods, services, deployments, and other Kubernetes components.
Kubernetes Eye is a Model Context Protocol (MCP) server that provides comprehensive management and diagnostic capabilities for Kubernetes clusters. It allows you to perform core Kubernetes operations, diagnose workload issues, and monitor resource usage directly through your AI assistant.
Clone the repository:
git clone https://github.com/wenhuwang/mcp-k8s-eye.git
cd mcp-k8s-eye
Build the binary:
go build -o mcp-k8s-eye
Make note of the full path to the binary for configuration.
Kubernetes Eye supports two modes of operation:
Add the following configuration to your AI assistant:
{
"mcpServers": {
"k8s eye": {
"command": "/path/to/your/mcp-k8s-eye",
"env": {
"HOME": "/your/home/directory"
}
}
}
}
Replace /path/to/your/mcp-k8s-eye with the actual path to the binary and /your/home/directory with your home directory where the kubeconfig file is located.
Start the MCP SSE server:
./mcp-k8s-eye serve --port 8080
Add the following configuration to your AI assistant:
{
"mcpServers": {
"k8s eye": {
"url": "http://localhost:8080/sse",
"env": {}
}
}
}
Once configured, you can interact with your Kubernetes cluster through your AI assistant. The tool provides various capabilities organized into three main categories:
Simply ask your AI assistant to perform Kubernetes operations, and it will use the appropriate tools from Kubernetes Eye to fulfill your requests.
Here are some example queries you can use:
The AI assistant will use the appropriate tools from Kubernetes Eye to respond to your queries with relevant information from your Kubernetes cluster.