MCP Directory

mcp-k8s-eye

Manage Kubernetes clusters and diagnose workload health via MCP.

Unverified
stdio (local)
No auth
Stale
Go
View repo 28

Add to your client

Copy the config for your MCP client and paste it into its config file.

Install / run
git clone https://github.com/wenhuwang/mcp-k8s-eye.git && cd mcp-k8s-eye && go build -o mcp-k8s-eye

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "mcp-k8s-eye": {
      "command": "YOUR mcp-k8s-eye PATH",
      "args": [],
      "env": {
        "HOME": "USER HOME DIR"
      }
    }
  }
}

Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf

Before you start

  • Go 1.23 or higher (to build the binary)
  • kubectl configured with access to a Kubernetes cluster
  • A valid kubeconfig file (referenced via the HOME environment variable)

About mcp-k8s-eye

mcp-k8s-eye is a Go-based MCP server that lets AI clients manage a Kubernetes cluster and analyze workload status. Beyond standard resource CRUD, pod exec/logs, and deployment scaling, it provides a suite of diagnostics tools that inspect pods, services, deployments, statefulsets, cronjobs, ingresses, network policies, admission webhooks, and nodes to surface configuration and health problems, plus workload-level CPU/memory usage. It supports Stdio and SSE transports.

Tools & capabilities (19)

resource_get

Get detailed resource information about a specific resource in a namespace.

resource_list

List detailed resource information about 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 a resource's detailed information 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 from a pod in a namespace.

pod_analyze

Diagnose all pods in a namespace (pod status, container status, resource utilization).

deployment_analyze

Diagnose all deployments in a namespace (available replicas).

statefulset_analyze

Diagnose all statefulsets in a namespace (service, pvc, available replicas).

service_analyze

Diagnose all services in a namespace (selector configuration, not-ready endpoints, events).

cronjob_analyze

Diagnose all cronjobs in a namespace (schedule, starting deadline, last schedule time).

ingress_analyze

Diagnose all ingresses in a namespace (ingress class, related services, tls secrets).

networkpolicy_analyze

Diagnose all network policies in a namespace (configuration, affected pods).

validatingwebhook_analyze

Diagnose all validating webhooks (configuration, referenced services and pods).

mutatingwebhook_analyze

Diagnose all mutating webhooks (configuration, referenced services and pods).

node_analyze

Diagnose all nodes in the cluster (node conditions).

workload_resource_usage

Get pod/deployment/replicaset/statefulset resource usage (cpu, memory) in a namespace.

What this server can do

mcp-k8s-eye provides tools for these capabilities — tap one to see every MCP server that does the same:

Deploy & manage cloud infra

When to use it

  • Inspect and manage Kubernetes resources (pods, deployments, services, CRDs) from an AI client
  • Diagnose unhealthy workloads by analyzing pod, deployment, and statefulset status
  • Troubleshoot networking and admission issues via ingress, network policy, and webhook diagnostics
  • Scale deployments, exec into pods, and fetch pod logs conversationally
  • Monitor workload CPU and memory usage across a namespace

Security notes

Operates against your configured Kubernetes cluster using your kubeconfig (set env.HOME to the directory containing the kubeconfig). It can create, update, and delete resources, scale deployments, and execute commands inside pods, so it has the same privileges as your kubeconfig context — restrict access accordingly.

mcp-k8s-eye FAQ

How do I install it?

Clone the repo and build the binary with Go 1.23+: 'git clone https://github.com/wenhuwang/mcp-k8s-eye.git && cd mcp-k8s-eye && go build -o mcp-k8s-eye'.

How does it connect to my cluster?

It uses your kubeconfig. In Stdio mode you set env.HOME to the directory containing the kubeconfig file.

Which transports are supported?

Both Stdio and SSE. For SSE, start the server and point the client at its /sse URL (e.g. http://localhost:8080/sse).

#kubernetes#k8s#devops#diagnostics#monitoring#cluster-management#golang

Alternatives to mcp-k8s-eye

Compare all alternatives →
GitHub MCP Server
DevOps & Cloud
31k

GitHub's official server for repos, issues, PRs, and Actions — local Docker or hosted remote.

Featured
Verified
stdio (local)
API key
Go
7 tools
Updated 15 days agoRepo
AWS API MCP Server
DevOps & Cloud
9.3k

Official AWS Labs server that lets agents call any AWS API via the AWS CLI surface.

Verified
stdio (local)
API key
Python
3 tools
Updated 17 days agoRepo
kubefwd
DevOps & Cloud
4.1k

Bulk Kubernetes port forwarding with an MCP server that lets AI assistants forward services, inspect pods, and monitor traffic.

Unverified
stdio (local)
No auth
Go
28 tools
Updated 9 days agoRepo

Compare mcp-k8s-eye with:

vs GitHub MCP Servervs AWS API MCP Servervs kubefwdvs Radar