mcp-k8s-eye

Purpose

mcp-k8s-eye is an MCP server for Kubernetes cluster management and application diagnostics. Its goal is to provide Kubernetes daily operations and cluster and application health analysis capabilities in the form of tools for AI clients to invoke.

Main Capabilities

Core Kubernetes Operations

Connect to a Kubernetes cluster.
General Kubernetes resource management capabilities, supporting all native resources, including Pod, Deployment, Service, StatefulSet, Ingress, etc., and supporting CustomResourceDefinition resources.
Resource operations include: list, get, create, update, delete.
Pod management capabilities: exec, logs.
Deployment management capabilities: scale.
Execute describe on Kubernetes resources.
Execute explain on Kubernetes resources.

Diagnostic Capabilities

Pod diagnostics: analyze Pod status, container status, and Pod resource utilization.
Service diagnostics: analyze Service selector configuration, unready endpoints, and events.
Deployment diagnostics: analyze the number of available replicas.
StatefulSet diagnostics: analyze whether the Service corresponding to the StatefulSet exists, whether the PVC exists, and the number of available replicas.
CronJob diagnostics: analyze the CronJob's schedule, starting deadline, and last schedule time.
Ingress diagnostics: analyze Ingress class configuration, related Services, and TLS Secrets.
NetworkPolicy diagnostics: analyze NetworkPolicy configuration and affected Pods.
ValidatingWebhook diagnostics: analyze webhook configuration, referenced Services, and Pods.
MutatingWebhook diagnostics: analyze webhook configuration, referenced Services, and Pods.
Node diagnostics: analyze Node conditions.
Cluster diagnostics and troubleshooting.

Monitoring Capabilities

Pod, Deployment, ReplicaSet, StatefulSet, DaemonSet workload resource usage (CPU, memory).
Node capacity and utilization (CPU, memory).
Cluster capacity and utilization (CPU, memory).

Advanced Features

Support multiple transport protocols: Stdio, SSE.
Support multiple AI clients.

Tool List

Resource Operation Tools

resource_get: Get detailed information of a specified resource in a namespace.
resource_list: List detailed information of all resources in a namespace.
resource_create_or_update: Create or update a resource in a namespace.
resource_delete: Delete a resource in a namespace.
resource_describe: Describe detailed information of a resource in a namespace.
deployment_scale: Scale a Deployment in a namespace.
pod_exec: Execute a command in a Pod in a namespace.
pod_logs: Get logs of a Pod in a namespace.

Diagnostic Tools

pod_analyze: Diagnose all Pods in a namespace.
deployment_analyze: Diagnose all Deployments in a namespace.
statefulset_analyze: Diagnose all StatefulSets in a namespace.
service_analyze: Diagnose all Services in a namespace.
cronjob_analyze: Diagnose all CronJobs in a namespace.
ingress_analyze: Diagnose all Ingresses in a namespace.
networkpolicy_analyze: Diagnose all NetworkPolicies in a namespace.
validatingwebhook_analyze: Diagnose all ValidatingWebhooks.
mutatingwebhook_analyze: Diagnose all MutatingWebhook.
node_analyze: Diagnose all Nodes in the cluster.

Monitoring Tools

workload_resource_usage: Get the resource usage (CPU, memory) of Pods, Deployments, ReplicaSets, and StatefulSets in a namespace.

Usage

Requires Go 1.23 or later.
Requires kubectl to be configured.
Use env.HOME to set the home directory where the kubeconfig file is located.
Start the MCP SSE server.
Configure your MCP server.

Applicable Scenarios

Perform daily operations such as resource queries, creation, updates, and deletion on a Kubernetes cluster through an AI client.
Perform diagnostic analysis on objects such as Pods, Deployments, Services, StatefulSets, CronJobs, Ingresses, NetworkPolicies, Webhooks, and Nodes.
View the CPU and memory usage of workloads, Nodes, and the cluster.

Limitations

The materials do not provide information on pricing, cost, platform support scope, performance, or effectiveness, and none is currently provided.