在本地磁盘以分类纯 Markdown 文件保存需要长期留存的事实,与智能体内置记忆并存。
记忆
vmware-pilot
试用为 VMware 多步操作设计审批流程、追踪状态、失败时自动回滚的统一编排入口。
它能做什么
Pilot 是 vmware-* 技能族的编排层。它接收自然语言目标,从 8 个伴随技能精选的 69 个工具中生成工作流草案,并在涉及破坏性操作前暂停等待人工审批。状态持久化在 SQLite 中,重启后可从最近检查点继续;步骤失败时已完成的步骤会按相反顺序回滚。多步、跨技能、需要审批的 VMware 场景用它;单台虚机操作、只读监控、负载均衡查询请分别走 vmware-aiops、vmware-monitor、vmware-avi。
什么时候用它
- 克隆虚机验证变更,通过审批后再合入生产
- 收到 Aria/NSX 告警后并行采集信号,并在检查点分流排查
- 先用 AVI 从池中排空流量,给服务器打补丁,再恢复入池
- 用自定义 YAML 模板串联 aiops + vks + avi,落地带 AKO Ingress 的应用部署
技能文档
VMware Pilot
Disclaimer: This is a community-maintained open-source project and is not affiliated with, endorsed by, or sponsored by VMware, Inc. or Broadcom Inc. "VMware" is a trademark of Broadcom. Source code is publicly auditable at github.com/vmware-skills/VMware-Pilot under the MIT license.
Multi-step workflow orchestration for VMware MCP skills — design, approve, execute, rollback.
Companion Skills: vmware-aiops (VM operations) | vmware-monitor (monitoring) | vmware-nsx (networking) | vmware-aria (metrics/alerts) | vmware-avi (load balancing/AKO)
What This Skill Does
| Capability | Description |
|---|---|
| Workflow Design | Natural language goal → AI designs steps from the get_skill_catalog building-block list (69 curated tools across 8 skills) |
| Approval Gates | Pause execution for human review before destructive operations |
| State Persistence | SQLite-backed, survives restarts, supports resume from checkpoint |
| Rollback | Reverse completed steps in order if workflow fails |
| Custom Templates | Save workflows as YAML for reuse, hot-reload without restart |
| Compliance Scans | Read-only health/capacity/anomaly checks across skills |
Quick Install
uv tool install vmware-pilot
vmware-pilot mcp # start the MCP server (stdio)
When to Use This Skill
| Scenario | Use Pilot? | Why |
|---|---|---|
| "Clone VM, test, then apply to prod" | Yes | Multi-step + approval |
| "Power on a VM" | No, use aiops | Single operation |
| "Set up app network + firewall + VMs" | Yes | Cross-skill orchestration |
| "Check cluster health" | No, use monitor/aria | Single read-only query |
| "Diagnose and fix an alert" | Yes | incident_response template |
| "Run compliance check" | Yes | compliance_scan template |
| "Drain server, patch, restore traffic" | Yes | Cross-skill: avi drain + aiops patch |
| "Deploy app with AKO ingress" | Yes | Cross-skill: aiops + vks + avi |
| "Check pool member health" | No, use avi | Single read-only query |
Related Skills — Skill Routing
| User Intent | Recommended Skill |
|---|---|
| VM lifecycle (power, clone, deploy) | vmware-aiops (uv tool install vmware-aiops) |
| Read-only monitoring | vmware-monitor (uv tool install vmware-monitor) |
| NSX networking (segments, gateways, NAT) | vmware-nsx (uv tool install vmware-nsx-mgmt) |
| NSX security (DFW, groups) | vmware-nsx-security (uv tool install vmware-nsx-security) |
| Aria metrics/alerts/capacity | vmware-aria (uv tool install vmware-aria) |
| Tanzu Kubernetes (Supervisor/TKC) | vmware-vks (uv tool install vmware-vks) |
| Storage (iSCSI, vSAN, datastores) | vmware-storage (uv tool install vmware-storage) |
| Load balancing, VS, pool, AKO | vmware-avi (uv tool install vmware-avi) |
| Audit log query | vmware-policy (vmware-audit CLI) |
| Multi-step orchestration | vmware-pilot (this skill) |
Common Workflows
1. Design a Custom Workflow (Interactive)
User: "I need to set up a new app environment with networking and VMs"
AI calls: get_skill_catalog() → see available tools
AI calls: design_workflow(goal="...") → create draft
AI calls: update_draft(id, steps=[...]) → fill in steps
User reviews and confirms
AI calls: confirm_draft(id, save_as_template=True)
AI calls: run_workflow(id) → execute with approval gates
2. Clone-and-Test (Built-in Template)
AI calls: plan_workflow("clone_and_test", {
target_vm: "db01",
change_spec: {memory_mb: 32768},
target: "vcenter-prod"
})
AI calls: run_workflow(workflow_id)
→ Clone → Apply → Monitor → [Approval Gate] → Commit → Cleanup
3. Batch Operations with Approval
AI calls: plan_workflow("plan_and_approve", {
operations: [
{action: "power_off", vm_name: "db01"},
{action: "revert_snapshot", vm_name: "db01", snapshot_name: "baseline"},
{action: "power_on", vm_name: "db01"}
]
})
→ Create Plan → [Approval Gate] → Execute Plan (with auto-rollback on failure)
4. Rolling Maintenance with AVI Drain
Drain traffic from a pool member via AVI, patch the server, then restore traffic:
1. vmware-avi pool disable # drain traffic from pool member
2. vmware-avi analytics # verify drain complete (0 active connections)
3. vmware-aiops vm guest-exec --cmd "apt-get upgrade -y" # patch the server
4. vmware-avi pool enable # restore traffic to pool member
5. vmware-avi pool members # verify health status is green
5. AKO-Aware Application Deployment
Deploy a backend VM, create a K8s namespace, and wire up AKO Ingress to the AVI Controller:
1. vmware-aiops deploy ova --name # deploy backend VM
2. vmware-vks namespace create # create K8s namespace
3. kubectl apply -f ingress.yaml # create Ingress with AKO annotations
4. vmware-avi ako ingress check # validate AKO annotations are correct
5. vmware-avi ako sync status # verify VS created on AVI Controller
Dispatch Contract (Important)
Pilot is a Dispatcher, not an Executor. It generates plans, tracks state, gates on approvals — it does NOT call companion skills' MCP tools itself. The calling AI agent is responsible for invoking vmware-aiops::vm_clone etc. when pilot's run_workflow returns a step description.
This is intentional v2-style architecture: pilot's context stays small, state is always on disk, and there are no persistent agent threads. Full contract details: see references/integration-patterns.md.
get_skill_catalog is a curated design aid, not a whitelist. It surfaces 69 hand-picked building blocks across 8 skills — a deliberate subset of what those skills expose (aiops alone has 49 tools; the catalog lists 18). A step's skill field is a free-form string handed to the calling agent, so a workflow may name any companion skill, including ones the catalog does not list — pilot itself (pilot) is used that way by built-in templates for approval gates. Use the catalog for inspiration; consult the target skill's own SKILL.md for its full tool surface.
MCP Tools (13 — 4 read, 9 write/control)
| Category | Tool | Risk | Description |
|---|---|---|---|
| Discovery | get_skill_catalog | low | Available skills and tools for design |
list_workflows | low | Built-in + custom templates | |
| Design | design_workflow | low | Natural language → draft |
update_draft | medium | Edit draft steps | |
confirm_draft | medium | Finalize draft → ready to execute | |
| Execute | plan_workflow | medium | Create from template |
create_workflow | medium | One-step custom creation | |
review_workflow | low | Structural sanity check before execution (approved | needs_revision) | |
run_workflow | medium | Execute next checkpoint (agent dispatches each step) | |
| Control | approve | high | Human approval to continue |
cancel_workflow | high | Cancel a workflow (approval rejected / unsafe) → terminal CANCELLED, can't be run | |
rollback | high | Reverse completed steps | |
get_workflow_status | low | State + audit log |
Built-in Templates (15)
The five most-used:
| Template | Steps | Approval | Skills Used |
|---|---|---|---|
clone_and_test | 6 | Yes | aiops + monitor |
incident_response | 4 | Yes | monitor + aiops |
investigate_alert | 4 / 8 | Yes | monitor + aria (parallel-group gather + 4-criteria checkpoint, optional deep_dive) |
plan_and_approve | 3 | Yes | aiops |
compliance_scan | 3 | No | monitor + aria |
Full list: clone_and_test, incident_response, investigate_alert, plan_and_approve, compliance_scan, network_segment_setup, vks_cluster_deploy, rolling_restart, capacity_expansion, disaster_recovery, patch_deployment, storage_expansion, baseline_capture, baseline_audit, baseline_remediate. See references/templates.md for full details.
Custom Templates
Drop YAML files in ~/.vmware/workflows/ — pilot auto-loads them.
# ~/.vmware/workflows/restart_cluster.yaml
name: restart_cluster
description: Rolling restart of database cluster
steps:
- action: check_health
skill: monitor
tool: get_alarms
params:
target: "{{target}}"
- action: stop_replica
skill: aiops
tool: vm_power_off
params:
vm_name: "{{replica_vm}}"
rollback_tool: vm_power_on
rollback_params:
vm_name: "{{replica_vm}}"
- action: require_approval
skill: pilot
tool: approve
params:
message: "Replica stopped. Proceed?"
- action: restart_primary
skill: aiops
tool: vm_power_off
params:
vm_name: "{{primary_vm}}"
Usage Mode
| Scenario | Recommended | Why |
|---|---|---|
| Local/small models (Ollama, Qwen) | MCP | Structured JSON I/O for multi-step state |
| Cloud models (Claude, GPT-4o) | MCP | Design mode needs structured tool calls |
| CI/CD pipeline orchestration | MCP | Programmatic plan/approve/run cycle |
| Quick template listing | MCP | Call list_workflows; the CLI has no template commands |
Note: every workflow operation — design, plan, run, approve, rollback — is MCP-only. The
vmware-pilotCLI exists to launch the server and report its version, nothing more. Other skills in the family (aiops, monitor, avi, etc.) offer full CLI and MCP modes.
CLI Quick Reference
The CLI is a launcher, not a second interface to workflows:
vmware-pilot mcp # start the MCP server (stdio)
vmware-pilot version # print installed version
vmware-pilot --help
# Validate a custom workflow YAML before loading
python3 scripts/validate_workflow.py ~/.vmware/workflows/my_workflow.yaml
# List available tools across all skills (design helper)
python3 scripts/list_available_tools.py # all skills
python3 scripts/list_available_tools.py aiops # specific skill
python3 scripts/list_available_tools.py --json # JSON output
# View audit logs (via vmware-policy)
vmware-audit log --last 20
vmware-audit log --status denied
Full CLI reference for companion skills: see
references/cli-reference.md
Troubleshooting
Workflow stuck in "awaiting_approval"
Call approve(workflow_id) with the correct workflow ID to continue, or rollback(workflow_id) to abort. If the MCP session was lost, reconnect and call get_workflow_status(workflow_id) to see the current state -- workflows persist in SQLite and survive restarts.
"Unknown workflow type" error from plan_workflow
The template name is case-sensitive. Use list_workflows() to see all available built-in and custom template names. Custom templates must be valid YAML in ~/.vmware/workflows/.
Custom YAML template not appearing
- Verify the file is in
~/.vmware/workflows/with a.yamlextension - Check YAML syntax -- run
python3 scripts/validate_workflow.pyto validate - Template names must be unique -- a custom template cannot shadow a built-in name
Rollback fails on some steps
Not all steps are reversible. Steps without rollback_tool defined are skipped during rollback. Pilot uses best-effort rollback: if one rollback step fails, it continues with remaining steps and reports which succeeded and which failed.
"Workflow cannot be run" state error
A workflow can only be run from pending or running states. If it is in draft, call confirm_draft() first. If it is in completed or failed, create a new workflow -- completed workflows cannot be re-run.
vmware-policy dependency missing
Pilot requires vmware-policy for the @vmware_tool decorator and audit logging. It is declared as a dependency in pyproject.toml and should install automatically. If missing, run pip install vmware-policy or reinstall pilot.
Setup
No vCenter credentials needed — pilot orchestrates other skills that handle connections.
{
"mcpServers": {
"vmware-pilot": {
"command": "vmware-pilot",
"args": ["mcp"]
}
}
}
Fallback:
{"command": "uvx", "args": ["--from", "vmware-pilot", "vmware-pilot-mcp"]}also works, butuvxre-resolves the package against PyPI on every start and fails behind a TLS-inspecting corporate proxy (invalid peer certificate: UnknownIssuer). The installed entry point above touches the network zero times; setUV_NATIVE_TLS=trueif you must useuvx.
Audit & Safety
All operations are automatically audited via vmware-policy (@vmware_tool decorator):
- Every tool call logged to
~/.vmware/audit.db(SQLite, framework-agnostic) - Policy rules enforced via
~/.vmware/rules.yaml(deny rules, maintenance windows, risk levels) - Risk classification: each tool tagged as low/medium/high/critical
- Environment scoping: policy rules can scope by environment (an optional label an opt-in
denyrule may match), and skills with a config may declareenvironment:per target. Pilot has no targets of its own and reports a constantlocal— its writes go to the local workflow DB, never to a VMware estate. Pilot's approval gate is a step in its own workflow: it pauses before the agent dispatches a destructive step, and the target skill then applies its own policy rules when the step runs - View recent operations:
vmware-audit log --last 20 - View denied operations:
vmware-audit log --status denied
vmware-policy is automatically installed as a dependency — no manual setup needed.
License
MIT
常见问题
- Pilot 会自己调用伴随技能吗?
- 不会。Pilot 只做调度——它返回步骤描述,由调用方代理去执行伴随技能的工具(如 vmware-aiops::vm_clone)。这样 Pilot 的上下文始终很小,状态始终落盘。
- MCP 会话断开后工作流怎么办?
- 状态保存在 SQLite。重新连接后调用 get_workflow_status(workflow_id) 查看当前状态,并从检查点继续。回滚是尽力而为:未配置 rollback_tool 的步骤会被跳过。
- 可以添加自定义模板吗?
- 可以。把 YAML 文件放到 ~/.vmware/workflows/,Pilot 会自动加载。上线前可用 scripts/validate_workflow.py 校验;自定义名称不能与内置模板重名。
相关技能
以 AI 机器人身份加入视频会议,提供语音、虚拟形象与屏幕共享四种模式。
把自然语言描述转为结构化 JSON,并由 mcp-diagram-generator MCP 服务生成 Draw.io、Mermaid 或 Excalidraw 图表文件。
诊断生产力系统反复失效的根因,给出最小干预——容量测算、瓶颈定位、可靠的本地记录。
从 AdMapix API 拉取广告创意、应用、榜单和收入预估等数据,原样返回结构化 JSON。
执行 Git 操作(提交、分支、合并、变基、冲突解决与恢复)时强制套用安全规则。
zw008 的更多技能
浏览全部技能通过 33 个 MCP 工具查询 VMware Aria Operations 的指标、告警、容量、异常与报告。
通过 21 个 MCP 工具或配套 CLI,管理 NSX-T 分布式防火墙策略、安全组、VM 标签、Traceflow 与 IDPS。
用 28 个读写工具管理 AVI 虚拟服务、AKO Ingress 与后端成员,覆盖 NSX ALB 运维场景。
用自然语言管理 vSphere Kubernetes Service 的 Supervisor 集群、命名空间与 TKC 集群生命周期。
通过 33 个 CLI/MCP 工具管理 NSX-T 网段、Tier-0/Tier-1 网关、NAT、静态路由与 IP 池。
针对 vSphere/ESXi/NSX 环境的合规扫描与基线比对,覆盖 CIS、STIG、等保 2.0 三级、PCI-DSS、IT-Grundschutz、NIS2。