Chin K

Kubernetes Agent Swarm

AI 智慧體v2.1.0
作者:Chin K更新於 2026-05-18共 11 個版本 · 最新版本釋出於 2026-04-15kubernetes@clawhub_kcns008/kubernetes
多 Agent 協作#Kubernetes#Multi Agent#Observability#Developer

據 AlphaMoat 統計,截至 2026-10,Kubernetes Agent Swarm 累計下載 30,662 次、獲 7 個星標,按累計下載量在本次參與排名的 10,914 個 AI 技能中排名第 897 位,在「AI 智慧體」分類(1,632 個)中排名第 139 位。

列表預設按累計下載量排序,也可按星標排序;詳情頁排名按累計下載量計算。

累計下載
3.1萬
星標
7
全站下載量排名
#897
全站前 9% · 「AI 智慧體」前 9%
平台累計下載星標評論
ClawHub ↗96766-
SkillHub ↗3.1萬7—

各平台的計數口徑不同,統計時間不一致。

這個技能是做什麼的

七個 agent:Orchestrator(Jarvis,任務路由與協調)、Cluster Ops(Atlas,叢集生命週期與節點)、GitOps(Flow,ArgoCD/Helm/Kustomize 部署)、Security(Shield,RBAC/策略/secrets/掃描)、Observability(Pulse,指標/日誌/告警)、Artifacts(Cache,映象倉庫/SBOM/CVE)、Developer Experience(Desk,名稱空間與支援)。這是純指令型技能:不包含可執行指令碼,agent 依據 markdown 指令呼叫主機上的 kubectl(必需)、oc/helm/jq(可選),需配置 KUBECONFIG。心跳按 5/10/15 分鐘分批排程;agent 間透過 @mention 協作並有升級路徑。能力邊界:可讀叢集狀態、經 GitOps 部署、排查事件;不可刪除生產資源、不可修改叢集級策略、不可直接改 secrets、不可執行不可逆升級,生產部署只能準備、由人批准。

資料釋出於 2026-10-27

可以用來做什麼

  • 用七個分工明確的 agent 協作完成平台運維
  • 讀叢集狀態、排查事件並準備 GitOps 部署
  • 透過 @mention 和升級路徑保持人機協同

什麼時候用

  • 平台團隊要統一協調 K8s/OpenShift 運維時
  • 需要 GitOps、安全、可觀測性分工協作時
  • 希望關鍵操作保留人工批准環節時

示例提示詞

以下是根據技能文件生成的示例,不代表實際執行結果。

@Pulse 幫我看看這個 CPU 飆升是釋出還是流量導致的。
@Atlas 檢查 staging 叢集的節點狀況。
總結一下當前叢集的健康檢查結果。

月度資料趨勢

累計下載3.1萬
1.6萬2.4萬3.2萬2026-09 · 累計下載 1.7萬2026-10 · 累計下載 3.1萬26-0926-10
星標7
6782026-09 · 星標 72026-10 · 星標 726-0926-10
月份累計下載環比星標
2026-103.1萬—7
2026-091.7萬—7

SKILL.md 原文

以下是作者釋出的原始檔案,按純文字顯示,本站未做修改或稽核;其中的連結不可點選,命令請核實後再執行。抓取於 2026-10-23。

5.8 KB · ClawHub v2.1.0
---
name: kubernetes
description: >
  Kubernetes & OpenShift Platform Agent Swarm — A coordinated multi-agent system for
  cluster operations. Includes Orchestrator (Jarvis), Cluster Ops (Atlas), GitOps (Flow),
  Security (Shield), Observability (Pulse), Artifacts (Cache), and Developer Experience (Desk).
  Pure instruction-based skill — no executable scripts.
metadata:
  author: kcns008
  version: 2.1.0
展開全部(共 181 行)
  agent_name: Swarm
  agent_role: Platform Agent Swarm (All Agents)
  session_key: "agent:platform:swarm"
  heartbeat: "*/5 * * * *"
  platforms:
    - openshift
    - kubernetes
    - eks
    - aks
    - gke
    - rosa
    - aro
  install_type: "instruction-only"
  always: false
  model_invocation: false
  requires:
    env:
      - KUBECONFIG
    binaries:
      - kubectl
  optional_binaries:
    - oc
    - helm
    - jq
  optional_env:
    - AWS_ACCESS_KEY_ID
    - AWS_SECRET_ACCESS_KEY
    - AZURE_CLIENT_ID
    - AZURE_CLIENT_SECRET
    - AZURE_TENANT_ID
    - GOOGLE_APPLICATION_CREDENTIALS
  credentials:
    - kubeconfig: "KUBECONFIG path or ~/.kube/config for cluster access"
    - cloud: "Optional cloud provider credentials for managed clusters (AWS/Azure/GCP)"
    - registry: "Optional container registry credentials for image operations"
---

# Kubernetes Agent Swarm — Platform Operations

A multi-agent system for Kubernetes and OpenShift platform operations. Seven specialized agents work together as a coordinated swarm.

## Runtime Requirements

| Requirement | Required | Description |
|-------------|----------|-------------|
| `kubectl` | ✅ Yes | Kubernetes CLI — must be in PATH |
| `oc` | Optional | OpenShift CLI — needed for OCP/ROSA/ARO |
| `helm` | Optional | For GitOps agent Helm operations |
| `jq` | Optional | For JSON output parsing |
| `KUBECONFIG` | ✅ Yes | Cluster access via env var or `~/.kube/config` |

Optional cloud CLIs (`aws`, `az`, `gcloud`, `rosa`) — only needed for managed cluster operations.

## Installation

```bash
clawhub install kubernetes
```

Or install individual agents:
```bash
clawhub install orchestrator
clawhub install cluster-ops
clawhub install gitops
clawhub install security
clawhub install observability
clawhub install artifacts
clawhub install developer-experience
```

## The Swarm — Agent Roster

| Agent | Code Name | Domain |
|-------|-----------|--------|
| Orchestrator | Jarvis | Task routing, coordination, standups |
| Cluster Ops | Atlas | Cluster lifecycle, nodes, upgrades |
| GitOps | Flow | ArgoCD, Helm, Kustomize, deploys |
| Security | Shield | RBAC, policies, secrets, scanning |
| Observability | Pulse | Metrics, logs, alerts, incidents |
| Artifacts | Cache | Registries, SBOM, promotion, CVEs |
| Developer Experience | Desk | Namespaces, onboarding, support |

## How It Works

This is an **instruction-only** skill. Agents receive markdown instructions describing what commands to run and how to interpret output. No executable scripts are included — the agent translates instructions into actions using the host's installed CLI tools.

### Session Setup

Before using the swarm, establish cluster context:

```bash
# Verify access
kubectl cluster-info
kubectl get nodes

# For OpenShift
oc status
```

### Agent Communication

Agents communicate via @mentions in shared task comments:
```
@Shield Please review the RBAC for payment-service v3.2 before I sync.
@Pulse Is the CPU spike related to the deployment or external traffic?
@Atlas The staging cluster needs 2 more worker nodes.
```

### Escalation Path
1. Agent detects issue
2. Agent attempts resolution within guardrails
3. If blocked → @mention another agent or escalate to human
4. P1 incidents → all relevant agents auto-notified

## Heartbeat Schedule

```
*/5  * * * *  Atlas, Pulse, Shield     (fast response: incidents, alerts, CVEs)
*/10 * * * *  Flow, Cache              (scheduled: deploys, promotions)
*/15 * * * *  Desk, Orchestrator       (batch: onboarding, standups)
```

## Agent Capabilities

### What Agents CAN Do
- Read cluster state (`kubectl get`, `kubectl describe`, `oc get`)
- Deploy via GitOps (`argocd app sync`, Flux reconciliation)
- Create documentation and reports
- Investigate and triage incidents
- Provision standard resources (namespaces, quotas, RBAC)
- Run health checks and audits
- Query metrics and logs

### What Agents CANNOT Do (Human-in-the-Loop Required)
- Delete production resources
- Modify cluster-wide policies
- Make direct changes to secrets without rotation workflow
- Perform irreversible cluster upgrades
- Approve production deployments (can prepare, human approves)

## Key Principles

- **Roles over genericism** — Each agent has a defined domain
- **Files over mental notes** — Only files persist between sessions
- **Human-in-the-loop** — Critical actions require approval
- **Guardrails over freedom** — Define what agents can and cannot do
- **Audit everything** — Every action logged

## File Structure

```
kubernetes/
├── SKILL.md                    # This file — combined swarm
├── AGENTS.md                   # Swarm configuration and protocols
├── skills/
│   ├── orchestrator/SKILL.md   # Jarvis — task routing
│   ├── cluster-ops/SKILL.md    # Atlas — cluster operations
│   ├── gitops/SKILL.md         # Flow — GitOps
│   ├── security/SKILL.md       # Shield — security
│   ├── observability/SKILL.md  # Pulse — monitoring
│   ├── artifacts/SKILL.md      # Cache — artifacts
│   └── developer-experience/SKILL.md  # Desk — DevEx
├── memory/MEMORY.md            # Long-term agent memory
├── working/WORKING.md          # Session progress
└── logs/LOGS.md                # Action audit trail
```

## Detailed Agent Documentation

See individual SKILL.md files for each agent's full capabilities, personality, and workflow instructions.

「AI 智慧體」分類下的熱門技能

資料月份:2026-10 · 下載/星標數聚合自公開技能註冊庫(ClawHub、SkillHub),按月更新。口徑說明見資料方法。