k8s-mcp-server MCP — 사용 사례, 설치 & 라이브 데모

왜 쓰나요

핵심 기능

Bundled CLIs: kubectl, helm, istioctl, argocd
Piping with jq/grep/sed for structured output
Two safety modes: restrictive (default) and permissive
Runs as non-root in a container — mount kubeconfig read-only
Configurable command timeout via K8S_MCP_TIMEOUT

라이브 데모

실제 사용 모습

k8s.replay ▶ 준비됨

0/0

설치

클라이언트 선택

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "k8s": {
      "command": "uvx",
      "args": [
        "k8s-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop → Settings → Developer → Edit Config 열기. 저장 후 앱 재시작.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "k8s": {
      "command": "uvx",
      "args": [
        "k8s-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Cursor는 Claude Desktop과 동일한 mcpServers 스키마 사용. 프로젝트 설정이 전역보다 우선.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "k8s": {
      "command": "uvx",
      "args": [
        "k8s-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Cline 사이드바의 MCP Servers 아이콘 클릭 후 "Edit Configuration" 선택.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "k8s": {
      "command": "uvx",
      "args": [
        "k8s-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop과 같은 형식. Windsurf 재시작 후 적용.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "k8s",
      "command": "uvx",
      "args": [
        "k8s-mcp-server"
      ]
    }
  ]
}

Continue는 맵이 아닌 서버 오브젝트 배열 사용.

~/.config/zed/settings.json

{
  "context_servers": {
    "k8s": {
      "command": {
        "path": "uvx",
        "args": [
          "k8s-mcp-server"
        ]
      }
    }
  }
}

context_servers에 추가. 저장 시 Zed가 핫 리로드.

claude mcp add k8s -- uvx k8s-mcp-server

한 줄 명령. claude mcp list로 확인, claude mcp remove로 제거.

사용 사례

실전 활용법: k8s-mcp-server

Morning cluster health check via chat

👤 SRE, platform engineers ⏱ ~5 min intermediate

언제 쓸까: Daily, before stand-up — anything red? Any pods crashlooping?

사전 조건

kubeconfig on host — Usually ~/.kube/config

흐름

Start the MCP container

Start k8s-mcp-server via Docker mounting ~/.kube read-only.✓ 복사됨

→ Container up
Run the checks

Across all namespaces, list pods not in Running state, nodes not Ready, and recent (last 1h) events of type Warning.✓ 복사됨

→ Three-part health report
Triage

For each crashlooping pod, get last 50 log lines and propose most likely cause.✓ 복사됨

→ Log excerpts + diagnoses

결과: 5-minute daily cluster pulse check.

함정

Running kubectl get events cluster-wide floods context — Always filter by time and type

Debug a failed Helm release

👤 Devs shipping Helm charts ⏱ ~15 min intermediate

언제 쓸까: helm upgrade failed and you want to know why without copying error messages back and forth.

흐름

Get history

helm history my-app -n production. What changed between v4 (working) and v5 (failed)?✓ 복사됨

→ Version diff
Diff templates

Render the template for v5 with current values and compare against v4. Highlight differences.✓ 복사됨

→ Diff of rendered manifests
Roll back if needed

Roll back to v4. Confirm all pods become Ready.✓ 복사됨

→ Rollback + health check

결과: Fast recovery with understanding, not just luck.

함정

Rolling back without capturing the failure first means you can't fix forward — Always save helm get all v5 to a file before rolling back

함께 쓰기: filesystem

Audit Argo CD apps for drift

👤 GitOps practitioners ⏱ ~20 min advanced

언제 쓸까: You suspect someone kubectl'd something into prod that Argo CD doesn't know about.

흐름

List out-of-sync apps

argocd app list --output json. Filter to apps where sync status isn't Synced.✓ 복사됨

→ Out-of-sync list
Diff each

For each drifted app, run argocd app diff and summarize what's different from Git.✓ 복사됨

→ Per-app diff summary

결과: Clear picture of manual drift to reconcile.

조합

다른 MCP와 조합해 10배 효율

k8s + github

When an app is broken in cluster, open GitHub issue with pod logs

If any pod in 'payments' namespace is crashlooping, fetch logs and open a GitHub issue with them attached.✓ 복사됨

도구

이 MCP가 노출하는 것

도구	입력	언제 호출	비용
kubectl	command: str (full kubectl args)	General cluster ops	free
helm	command: str	Release management	free
istioctl	command: str	Service mesh ops	free
argocd	command: str	GitOps ops (requires argocd login in container)	free

비용 및 제한

운영 비용

API 쿼터: None
호출당 토큰: kubectl get -o yaml can be huge; prefer -o json | jq to trim
금액: Free, MIT
팁: Always use -o wide or -o jsonpath=... to trim. kubectl get all is a token-bomb.

보안

권한, 시크릿, 파급범위

최소 스코프: Kubeconfig scoped to a least-privilege ServiceAccount

자격 증명 저장: Mount ~/.kube read-only into the container

데이터 외부 송신: Traffic to your cluster API server only

절대 부여 금지: cluster-admin ServiceAccount to the kubeconfig used by the MCP

Default mode is restrictive — whitelists commands. Permissive mode allows anything. Don't switch to permissive unless you trust the prompt chain.
Always review delete, scale --replicas=0, drain, and apply commands before confirming.

문제 해결

자주 발생하는 오류와 해결

The connection to the server localhost:8080 was refused

kubeconfig not mounted or KUBECONFIG env var wrong in the container.

확인: docker exec ... kubectl config view

Command blocked in restrictive mode

Either craft a narrower command, or explicitly opt into permissive mode for that session.

Timeout on long-running command

Raise K8S_MCP_TIMEOUT. For truly long ops, run async via a Job and poll.

대안

k8s-mcp-server 다른 것과 비교

대안	언제 쓰나	단점/장점
Direct kubectl via ssh-manager	You prefer running kubectl on a jump host	Less structured; no command validation
Kagent / AIOps SaaS MCPs	You want managed service rather than self-hosted	Costs money; less control

k8s-mcp-server