Vmware Monitor

VMware Monitor (Read-Only)

Safe, read-only VMware vCenter and ESXi monitoring skill. Query your entire VMware infrastructure using natural language through any AI coding assistant — without risk of accidental modifications.

Code-level safety: This is an independent repository (zw008/VMware-Monitor). No destructive code paths exist — no power off, delete, create, reconfigure, snapshot-create/revert/delete, clone, or migrate functions are present in the codebase. For full operations, use the separate VMware-AIops repo. Install: clawhub install vmware-aiops

When to Use This Skill

• Query VM, host, datastore, cluster, and network inventory
• Check health status, active alarms, hardware sensors, and event logs
• View VM details and existing snapshot lists (read-only)
• Run scheduled log scanning with webhook notifications (Slack, Discord)
• You need zero-risk monitoring — no accidental power-off, delete, or reconfigure possible

Quick Install

Works with Claude Code, Cursor, Codex, Gemini CLI, Trae, Kimi, and 30+ AI agents:

# Via Skills.sh
npx skills add zw008/VMware-Monitor

# Via ClawHub
clawhub install vmware-monitor

Claude Code

/plugin marketplace add zw008/VMware-Monitor
/plugin install vmware-monitor
/vmware-monitor:vmware-monitor

Usage Mode

Choose the best mode based on your AI tool:

Calling Priority

• MCP-native tools (Claude Code, Cursor): MCP first, CLI fallback
• All other tools: CLI first (MCP not needed)

Tip: If your AI tool supports MCP, check whether vmware-monitor MCP server is loaded (/mcp in Claude Code). If not, configure it first — MCP provides the best hands-free experience.

CLI Examples

# Activate venv first
source /path/to/VMware-Monitor/.venv/bin/activate

# Inventory
vmware-monitor inventory vms --target home-esxi
vmware-monitor inventory hosts --target home-esxi

# Health
vmware-monitor health alarms --target home-esxi
vmware-monitor health events --hours 24 --severity warning --target home-esxi

# VM info (read-only)
vmware-monitor vm info my-vm --target home-esxi

MCP Mode (Optional)

For Claude Code / Cursor users who prefer structured tool calls, add to ~/.claude/settings.json:

{
  "mcpServers": {
    "vmware-monitor": {
      "command": "/path/to/VMware-Monitor/.venv/bin/python",
      "args": ["-m", "mcp_server"],
      "cwd": "/path/to/VMware-Monitor",
      "env": {
        "VMWARE_MONITOR_CONFIG": "~/.vmware-monitor/config.yaml"
      }
    }
  }
}

MCP exposes 7 read-only tools: list_virtual_machines, list_esxi_hosts, list_all_datastores, list_all_clusters, get_alarms, get_events, vm_info. All accept optional target parameter.

list_virtual_machines supports limit, sort_by, power_state, fields for compact context in large inventories.

Architecture

User (Natural Language)
  ↓
AI Tool (Claude Code / Aider / Gemini / Codex / Cursor / Trae / Kimi)
  ↓
  ├─ CLI mode (default): vmware-monitor CLI ──→ pyVmomi ──→ vSphere API
  │
  └─ MCP mode (optional): MCP Server (stdio) ──→ pyVmomi ──→ vSphere API
  ↓
vCenter Server ──→ ESXi Clusters ──→ VMs
    or
ESXi Standalone ──→ VMs

First Interaction: Environment Selection

When the user starts a conversation, always ask first:

1. Which environment do they want to monitor? (vCenter Server or standalone ESXi host)
2. Which target from their config? (e.g., prod-vcenter, lab-esxi)
3. If no config exists yet, guide them through creating ~/.vmware-monitor/config.yaml

Capabilities (Read-Only)

1. Inventory

2. Health & Monitoring

Monitored Event Types:

3. VM Info & Snapshot List (Read-Only)

4. Scheduled Scanning & Notifications

FORBIDDEN Operations — DO NOT EXIST IN CODEBASE

These operations cannot be performed with this skill — zero destructive code paths exist:

• ❌ vm power-on/off, vm reset, vm suspend
• ❌ vm create/delete/reconfigure
• ❌ vm snapshot-create/revert/delete
• ❌ vm clone/migrate

Direct users to VMware-AIops (clawhub install vmware-aiops) for these.

Safety Features

Version Compatibility

pyVmomi auto-negotiates the API version during SOAP handshake — no manual configuration needed.

Supported AI Platforms

MCP Server — Local Agent Compatibility

The MCP server works with any MCP-compatible agent via stdio transport. All 8 tools are read-only. Config templates in examples/mcp-configs/:

# Example: Aider + Ollama (fully local, no cloud API)
aider --conventions codex-skill/AGENTS.md --model ollama/qwen2.5-coder:32b

CLI Reference

# Diagnostics
vmware-monitor doctor [--skip-auth]

# MCP Config Generator
vmware-monitor mcp-config generate --agent <goose|cursor|claude-code|continue|vscode-copilot|localcowork|mcp-agent>
vmware-monitor mcp-config list

# Inventory
vmware-monitor inventory vms [--target <name>] [--limit <n>] [--sort-by name|cpu|memory_mb|power_state] [--power-state poweredOn|poweredOff]
vmware-monitor inventory hosts [--target <name>]
vmware-monitor inventory datastores [--target <name>]
vmware-monitor inventory clusters [--target <name>]

# Health
vmware-monitor health alarms [--target <name>]
vmware-monitor health events [--hours 24] [--severity warning]

# VM Info (read-only)
vmware-monitor vm info <vm-name>
vmware-monitor vm snapshot-list <vm-name>

# Scanning & Daemon
vmware-monitor scan now [--target <name>]
vmware-monitor daemon start
vmware-monitor daemon stop
vmware-monitor daemon status

Setup

# 1. Install via uv (recommended) or pip
uv tool install vmware-monitor
# Or: pip install vmware-monitor

# 2. Configure
mkdir -p ~/.vmware-monitor
vmware-monitor init  # generates config.yaml and .env templates
chmod 600 ~/.vmware-monitor/.env
# Edit ~/.vmware-monitor/config.yaml and .env with your target details

Development Install

git clone https://github.com/zw008/VMware-Monitor.git
cd VMware-Monitor
uv venv && source .venv/bin/activate
uv pip install -e .

Security

• Read-Only by Design: This is an independent repository with zero destructive code paths. No power off, delete, create, reconfigure, or migrate functions exist in the codebase.
• Source Code: Fully open source at github.com/zw008/VMware-Monitor. The uv installer (vmware-monitor) installs from this repository. We recommend reviewing the source code and commit history before deploying in production.
• TLS Verification: Enabled by default. The disableSslCertValidation option exists solely for ESXi hosts using self-signed certificates (common in home labs). In production, always use CA-signed certificates with full TLS verification.
• Config File Contents: ~/.vmware-monitor/config.yaml stores target hostnames, ports, and a reference to the .env file. It does not contain passwords or tokens. All secrets (vCenter username/password) are stored exclusively in ~/.vmware-monitor/.env (chmod 600), loaded via python-dotenv. We recommend using a least-privilege read-only vCenter service account.
• Webhook Data Scope: Webhook notifications are disabled by default. When enabled, they send monitoring summaries (alarm counts, event types, host status) to user-configured URLs only (Slack, Discord, or any HTTP endpoint you control). No data is sent to third-party services. Webhook payloads contain no credentials, IPs, or personally identifiable information — only aggregated alert metadata.
• Prompt Injection Protection: All vSphere-sourced content (event messages, host logs) is truncated, stripped of control characters, and wrapped in boundary markers ([VSPHERE_EVENT]/[VSPHERE_HOST_LOG]) before output to prevent prompt injection when consumed by LLM agents.

License

MIT