{
  "markdown": "<!-- mcp-name: io.github.vmware-skills/vmware-aiops -->\n# VMware AIops\n\n> **Author**: Wei Zhou, VMware by Broadcom — wei-wz.zhou@broadcom.com\n> This is a community-driven project by a VMware engineer, not an official VMware product.\n> For official VMware developer tools see [developer.broadcom.com](https://developer.broadcom.com).\n\nEnglish | [中文](README-CN.md)\n\nAI-powered VMware vCenter/ESXi VM lifecycle and deployment tool — 60 tools.\n\n> **Companion skills** handle everything else:\n>\n> | Skill | Scope | Install |\n> |-------|-------|---------|\n> | **[vmware-monitor](https://github.com/vmware-skills/VMware-Monitor)** | Read-only: inventory, health, alarms, events, metrics | `uv tool install vmware-monitor` |\n> | **[vmware-storage](https://github.com/vmware-skills/VMware-Storage)** | Datastores, iSCSI, vSAN management | `uv tool install vmware-storage` |\n> | **[vmware-vks](https://github.com/vmware-skills/VMware-VKS)** | Tanzu Namespaces, TKC cluster lifecycle | `uv tool install vmware-vks` |\n>\n> **Need read-only monitoring only?** Use [VMware-Monitor](https://github.com/vmware-skills/VMware-Monitor) — zero destructive code in the codebase.\n\n[![ClawHub](https://img.shields.io/badge/ClawHub-vmware--aiops-orange)](https://clawhub.ai/skills/vmware-aiops)\n[![Skills.sh](https://img.shields.io/badge/Skills.sh-Install-blue)](https://skills.sh/vmware-skills/VMware-AIops)\n[![Claude Code Marketplace](https://img.shields.io/badge/Claude_Code-Marketplace-blueviolet)](https://github.com/vmware-skills/VMware-AIops)\n[![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](LICENSE)\n\n## ⚡ Quick Investigation Reports (read-only)\n\nTriage → investigate → act, all in one conversation. Five opinionated read-only reports **aggregate and correlate server-side** and hand back a high-signal result (never raw inventory), so you can decide *where to look* before changing anything. Each renders a **self-contained offline HTML snapshot** with `--html` (no external assets; drill-down detail collapses in native `<details>`, zero JavaScript). All delegate to the [vmware-monitor](https://github.com/vmware-skills/VMware-Monitor) library using AIops's own vCenter connection.\n\n| Question | Command | What it correlates |\n|----------|---------|--------------------|\n| **\"What needs attention now?\"** across all vCenters | `vmware-aiops attention` | Every vCenter merged into one globally-ranked issue list; unreachable targets degrade gracefully |\n| **\"Is anything on fire?\"** across all clusters | `vmware-aiops summary` | Every cluster's hosts + VM power + live CPU/mem + alarms → ranked top-N issues + per-cluster status |\n| **\"What's happening around this VM?\"** | `vmware-aiops investigate vm <name>` | VM state + host + cluster + backing datastores + snapshots + alarms + performance + a merged event timeline |\n| **\"What's happening around this host?\"** | `vmware-aiops investigate host <name>` | Host state + cluster + the VMs it runs + mounted datastores + alarms + performance + correlated timeline |\n| **\"What's happening around this datastore?\"** | `vmware-aiops investigate datastore <name>` | Capacity/free + mounting hosts + VMs it backs + alarms + correlated timeline |\n\n```bash\nvmware-aiops attention                            # what needs attention now, all vCenters\nvmware-aiops investigate vm web-01 --hours 72     # everything around a VM, then act on it\nvmware-aiops investigate vm web-01 --html         # → offline snapshot in ~/vmware-health/\n```\n\nVia MCP these are the tools `cluster_health_summary`, `cross_vcenter_attention`, `vm_investigation_bundle`, `host_investigation_bundle`, `datastore_investigation_bundle`. (Requires `vmware-monitor` installed.)\n\n### Quick Install (Recommended)\n\nWorks with Claude Code, Cursor, Codex, Gemini CLI, Trae, and 30+ AI agents:\n\n```bash\n# Via Skills.sh\nnpx skills add vmware-skills/VMware-AIops\n\n# Via ClawHub\nclawhub install @zw008/vmware-aiops\n```\n\n### PyPI Install (No GitHub Access Required)\n\n```bash\n# Install via uv (recommended)\nuv tool install vmware-aiops\n\n# Or via pip\npip install vmware-aiops\n\n# China mainland mirror (faster)\npip install vmware-aiops -i https://pypi.tuna.tsinghua.edu.cn/simple\n```\n\n### Offline / Air-Gapped Install (from source)\n\nThis project uses the modern PEP 517 build system (hatchling), so there is **no\n`setup.py`** by design — that is expected, not a missing file. If you cloned the\nsource and hit `ERROR: File \"setup.py\" or \"setup.cfg\" not found ... editable mode\ncurrently requires a setuptools-based build`, your `pip` is older than 21.3 and\ncannot do an *editable* (`-e`) install with a non-setuptools backend. Editable\nmode is a developer convenience, not needed to run the tool — do one of:\n\n```bash\n# From the source tree — a normal (non-editable) install builds a wheel:\npip install .              # NOT  pip install -e .\n\n# ...or upgrade pip first, and editable works too:\npip install --upgrade pip && pip install -e .\n```\n\nFor a **truly air-gapped host**, build the wheels on a connected machine and copy\nthem over — the target then needs no network:\n\n```bash\n# On a connected machine, collect this package + its dependencies as wheels:\npip wheel . -w dist        # → dist/*.whl   (or: uv build, for just this package)\n\n# Copy dist/ to the air-gapped host, then install offline:\npip install --no-index --find-links dist vmware-aiops\n```\n\n---\n\n## Why this over other VMware MCP servers\n\nMost open-source VMware MCP servers (e.g. `bright8192/esxi-mcp-server`,\n`giuliolibrando/vmware-vsphere-mcp-server`) are **single-vCenter VM wrappers**:\nlist/power/snapshot a VM, basic monitoring, a `confirm=True` flag. They explicitly\ndo not cover networking, storage, Kubernetes, ops analytics, load balancing, or\ncompliance — and \"logging is documented\" is not an audit trail.\n\nThis is one skill in an **11-package family** that covers the whole estate and runs\nevery tool through a governed harness:\n\n| | Other VMware MCP servers | This family |\n|---|:---:|:---:|\n| VM lifecycle + monitoring | ✅ | ✅ |\n| NSX networking (segments/gateways/NAT/routing/IPAM) | ❌ | ✅ vmware-nsx |\n| NSX security (DFW/groups/IDS-IPS/traceflow) | ❌ | ✅ vmware-nsx-security |\n| Storage (datastore/iSCSI/vSAN) | ❌ | ✅ vmware-storage |\n| Tanzu Kubernetes (Supervisor/Namespace/TKC) | ❌ | ✅ vmware-vks |\n| Aria Operations (metrics/alerts/capacity) | ❌ | ✅ vmware-aria |\n| AVI / NSX ALB load balancing + AKO | ❌ | ✅ vmware-avi |\n| Compliance baselines + drift (CIS/SCG/等保/PCI) | ❌ | ✅ vmware-harden |\n| **Governed harness** (unified audit, policy engine, token budget + runaway breaker, graduated risk tiers, undo-token, prompt-injection sanitize) | ❌ | ✅ vmware-policy on every tool |\n\nIf you only ever power-cycle VMs in one vCenter, a single-file server is fine. If you\nrun a real (regulated, NSX-segmented, multi-domain) VMware estate and need an AI\noperator an auditor can sign off on, that's what this family is for — see\n[docs/compliance-ready.md](../docs/compliance-ready.md).\n\n## Capabilities Overview\n\n### What This Skill Does\n\n| Category | Tools | Count |\n|----------|-------|:-----:|\n| **VM Lifecycle** | power on/off, TTL auto-delete, clean slate | 6 |\n| **Deployment** | OVA, template, linked clone, batch clone/deploy | 8 |\n| **Guest Ops** | exec commands, upload/download files, provision | 5 |\n| **Plan/Apply** | multi-step planning with rollback | 4 |\n| **Cluster** | create, delete, HA/DRS config, add/remove hosts | 6 |\n| **Datastore** | browse files, scan for images | 2 |\n| **Network** | dvSwitch portgroup list/create, host VMkernel list/add/remove, DF-bit MTU-path ping | 6 |\n\n### CLI vs MCP: Which Mode to Use\n\n| Scenario | Recommended | Why |\n|----------|:-----------:|-----|\n| **Local/small models** (Ollama, Qwen <32B) | **CLI** | ~2K tokens context vs ~10K for MCP; small models struggle with many tool schemas |\n| **Token-sensitive workflows** | **CLI** | SKILL.md + Bash tool = minimal overhead |\n| **Cloud models** (Claude, GPT-4o) | Either | Both work; MCP gives structured JSON I/O |\n| **Automated pipelines / Agent chaining** | **MCP** | Type-safe parameters, structured output, no shell parsing |\n| **Monitoring / storage / K8s** | Companion skills | See [vmware-monitor](https://github.com/vmware-skills/VMware-Monitor), [vmware-storage](https://github.com/vmware-skills/VMware-Storage), [vmware-vks](https://github.com/vmware-skills/VMware-VKS) |\n\n> **Rule of thumb**: Use CLI for cost efficiency and small models. Use MCP for structured automation with large models.\n\n### Architecture\n\n```\nUser (Natural Language)\n  ↓\nAI CLI Tool (Claude Code / Gemini / Codex / Aider / Continue / Trae / Kimi)\n  ↓ reads SKILL.md / AGENTS.md / rules\n  ↓\nvmware-aiops CLI\n  ↓ pyVmomi (vSphere SOAP API)\n  ↓\nvCenter Server ──→ ESXi Cluster ──→ VM\n    or\nESXi Standalone Host ──→ VM\n```\n\n### Version Compatibility\n\n| vSphere / VCF Version | Support | Notes |\n|----------------|---------|-------|\n| VCF 9.1 / vSphere 9.1 | ✅ Full | Released 2026-05-12. pyVmomi `<10.0` resolves and connects via SOAP; new REST-only features (`PATCH /deployment/size`, IPv6-only GOSC) not yet wrapped — see [VCF Python SDK](https://developer.broadcom.com/sdks) for those. |\n| VCF 9.0 / vSphere 9.0 | ✅ Full | pyVmomi 8.0.3+ connects against vSphere 9 SOAP API. From VCF 9, pyVmomi is also bundled inside the unified VCF Python SDK. |\n| 8.0 / 8.0U1-U3 | ✅ Full | `CreateSnapshot_Task` deprecated → use `CreateSnapshotEx_Task` |\n| 7.0 / 7.0U1-U3 | ✅ Full | All APIs supported |\n| 6.7 | ✅ Compatible | Backward-compatible, tested |\n| 6.5 | ✅ Compatible | Backward-compatible, tested |\n\n> pyVmomi auto-negotiates the API version during SOAP handshake — no manual configuration needed. The same codebase manages 7.0 / 8.0 / 9.0 / 9.1 environments seamlessly.\n\n#### Official Broadcom References\n\n- **SDKs**: <https://developer.broadcom.com/sdks> — VCF Python SDK (recommended for VCF 9+, bundles pyVmomi + vSAN SDK), vSphere Automation SDK for Python (REST APIs)\n- **REST APIs**: <https://developer.broadcom.com/xapis> — vSphere Automation API, VCF API, SDDC Manager API\n- **CLI Tools**: <https://developer.broadcom.com/tools> — PowerCLI 9.1, ESXCLI, OVF Tool\n\n---\n\n## Common Workflows\n\n### Deploy a Lab Environment\n\n1. Browse datastore for OVA images → `vmware-aiops datastore browse <ds> --pattern \"*.ova\"`\n2. Deploy VM from OVA → `vmware-aiops deploy ova ./image.ova --name lab-vm --datastore ds1`\n3. Install software inside VM → `vmware-aiops vm guest-exec lab-vm --cmd /bin/bash --args \"-c 'apt-get install -y nginx'\" --user root`\n4. Create baseline snapshot → `vmware-aiops vm snapshot-create lab-vm --name baseline`\n5. Set TTL for auto-cleanup → `vmware-aiops vm set-ttl lab-vm --minutes 480`\n\n### Batch Clone for Testing\n\n1. Create plan: `vm_create_plan` with multiple clone + reconfigure steps\n2. Review plan with user (shows affected VMs, irreversible warnings)\n3. Apply: `vm_apply_plan` executes sequentially, stops on failure\n4. If failed: `vm_rollback_plan` reverses executed steps\n5. Set TTL on all clones for auto-cleanup\n\n### Migrate VM to Another Host\n\n1. Check VM info via `vmware-monitor` → verify power state and current host\n2. Migrate: `vmware-aiops vm migrate my-vm --to-host esxi-02`\n3. Verify migration completed\n\n---\n\n## VM Lifecycle\n\n| Operation | Command | Confirmation | vCenter | ESXi |\n|-----------|---------|:------------:|:-------:|:----:|\n| Power On | `vm power-on <name>` | — | ✅ | ✅ |\n| Graceful Shutdown | `vm power-off <name>` | Double | ✅ | ✅ |\n| Force Power Off | `vm power-off <name> --force` | Double | ✅ | ✅ |\n| Reset | `vm reset <name>` | — | ✅ | ✅ |\n| Suspend | `vm suspend <name>` | — | ✅ | ✅ |\n| Create VM | `vm create <name> --cpu --memory --disk` | — | ✅ | ✅ |\n| Delete VM | `vm delete <name>` | Double | ✅ | ✅ |\n| Reconfigure | `vm reconfigure <name> --cpu --memory` | Double | ✅ | ✅ |\n| Create Snapshot | `vm snapshot-create <name> --name <snap>` | — | ✅ | ✅ |\n| List Snapshots | `vm snapshot-list <name>` | — | ✅ | ✅ |\n| Revert Snapshot | `vm snapshot-revert <name> --name <snap>` | Double | ✅ | ✅ |\n| Delete Snapshot | `vm snapshot-delete <name> --name <snap> [--no-wait]` | Double | ✅ | ✅ |\n| Task Status | `vm task-status <task-id>` | — | ✅ | ✅ |\n| Clone VM | `vm clone <name> --new-name <new>` | Double | ✅ | ✅ |\n| vMotion | `vm migrate <name> --to-host <host>` | Double | ✅ | ❌ |\n| **Set TTL** | `vm set-ttl <name> --minutes <n>` | Double | ✅ | ✅ |\n| **Cancel TTL** | `vm cancel-ttl <name>` | — | ✅ | ✅ |\n| **List TTLs** | `vm list-ttl` | — | ✅ | ✅ |\n| **Clean Slate** | `vm clean-slate <name> [--snapshot baseline]` | Double | ✅ | ✅ |\n| **Guest Exec** | `vm guest-exec <name> --cmd /bin/bash --args \"...\"` | Double | ✅ | ✅ |\n| **Guest Exec (with output)** | `vm guest-exec-output <name> --cmd \"df -h\"` | — | ✅ | ✅ |\n| **Guest Upload** | `vm guest-upload <name> --local f.sh --guest /tmp/f.sh` | Double | ✅ | ✅ |\n| **Guest Download** | `vm guest-download <name> --guest /var/log/syslog --local ./syslog` | — | ✅ | ✅ |\n\n> Guest Operations require VMware Tools running inside the guest OS. `guest-exec-output` auto-detects Linux/Windows shell and captures stdout/stderr.\n\n### Plan → Apply (Multi-step Operations)\n\nFor complex operations involving 2+ steps or 2+ VMs, use the plan/apply workflow instead of executing individually:\n\n| Step | What Happens |\n|------|-------------|\n| 1. **Create Plan** | AI calls `vm_create_plan` — validates actions, checks targets in vSphere, generates plan with rollback info |\n| 2. **Review** | AI shows plan to user: steps, affected VMs, irreversible warnings |\n| 3. **Apply** | `vm_apply_plan` executes sequentially; stops on failure |\n| 4. **Rollback** (if failed) | Asks user whether to rollback, then `vm_rollback_plan` reverses executed steps (irreversible steps skipped) |\n\nPlans stored in `~/.vmware-aiops/plans/`, auto-deleted on success, auto-cleaned after 24h.\n\n## VM Deployment & Provisioning\n\n| Operation | Command | Speed | vCenter | ESXi |\n|-----------|---------|:-----:|:-------:|:----:|\n| Deploy from OVA | `deploy ova <path> --name <vm>` | Minutes | ✅ | ✅ |\n| Deploy from Template | `deploy template <tmpl> --name <vm>` | Minutes | ✅ | ✅ |\n| Linked Clone | `deploy linked-clone --source <vm> --snapshot <snap> --name <new>` | Seconds | ✅ | ✅ |\n| Attach ISO | `deploy iso <vm> --iso \"[ds] path/to.iso\"` | Instant | ✅ | ✅ |\n| Convert to Template | `deploy mark-template <vm>` | Instant | ✅ | ✅ |\n| Batch Clone | `deploy batch-clone --source <vm> --count <n>` | Minutes | ✅ | ✅ |\n| Batch Deploy (YAML) | `deploy batch spec.yaml` | Auto | ✅ | ✅ |\n\n## Cluster Management\n\n| Operation | Command | Confirmation | vCenter | ESXi |\n|-----------|---------|:------------:|:-------:|:----:|\n| Cluster Info | `cluster info <name>` | — | ✅ | ❌ |\n| Create Cluster | `cluster create <name> [--ha] [--drs]` | — | ✅ | ❌ |\n| Delete Cluster | `cluster delete <name>` | Double | ✅ | ❌ |\n| Add Host | `cluster add-host <cluster> --host <host>` | Double | ✅ | ❌ |\n| Remove Host | `cluster remove-host <cluster> --host <host>` | Double | ✅ | ❌ |\n| Configure HA/DRS | `cluster configure <name> [--ha/--no-ha] [--drs/--no-drs]` | Double | ✅ | ❌ |\n\n> `remove-host` requires the host to be in **maintenance mode** first; the host is moved out of the cluster into the datacenter's host folder as a standalone host.\n\n## Alarm Management\n\n| Operation | Command | Confirmation | vCenter | ESXi |\n|-----------|---------|:------------:|:-------:|:----:|\n| List Triggered Alarms | `alarm list [--target <t>]` | — | ✅ | ❌ |\n| Acknowledge Alarm | `alarm acknowledge <entity> <alarm>` | — | ✅ | ❌ |\n| Clear (Reset) Alarms | `alarm reset <entity> <alarm>` | Double | ✅ | ❌ |\n\n> **Blast radius**: vSphere has no per-alarm clear API. `alarm reset` uses `AlarmManager.ClearTriggeredAlarms`, which clears **all** triggered alarms matching the named alarm's entity type (host/VM/all) and current status (red/yellow) — not just the named one. The named alarm is looked up first (typos fail fast), and the output's `scope` field reports exactly what was cleared. Cleared alarms re-trigger automatically if their underlying condition persists.\n\n## Datastore Browser\n\n| Feature | vCenter | ESXi | Details |\n|---------|:-------:|:----:|---------|\n| Browse Files | ✅ | ✅ | List files/folders in any datastore path |\n| Scan Images | ✅ | ✅ | Discover ISO, OVA, OVF, VMDK across all datastores |\n\n## Scheduled Scanning & Notifications\n\n| Feature | Details |\n|---------|---------|\n| Daemon | APScheduler-based, configurable interval (default 15 min) |\n| Multi-target Scan | Sequentially scan all configured vCenter/ESXi targets |\n| Scan Content | Alarms + Events + Host logs (hostd, vmkernel, vpxd) |\n| Log Analysis | Regex pattern matching: error, fail, critical, panic, timeout, corrupt |\n| Structured Log | JSONL output to `~/.vmware-aiops/scan.log` |\n| Webhook | Slack, Discord, or any HTTP endpoint |\n| Daemon Management | `daemon start/stop/status`, PID file, graceful shutdown |\n\n## Safety Features\n\n| Feature | Details |\n|---------|---------|\n| **Dry-Run Mode** (**CLI only**) | `--dry-run` on any destructive CLI command prints exact API calls without executing |\n| **Plan → Confirm → Execute → Log** | CLI workflow: show current state, confirm changes, execute, audit log |\n| **Double Confirmation** (**CLI only**) | Destructive CLI commands (power-off, delete, reconfigure, snapshot-revert/delete, clean-slate, guest-exec, guest-upload, cluster delete/remove-host, alarm clear) require 2 sequential prompts and take no bypass flag |\n| **No confirmation on the MCP path** | The 43 write tools an agent sees over MCP act on the first call — no `confirmed=` handshake, no approval tier, no read-only switch. What decides whether a write lands is the privilege of the vCenter account, and what records it is the audit trail. See [What protects you](#what-protects-you) |\n| **Rejection Logging** | Declined CLI confirmations are recorded in the audit trail |\n| **Audit Trail** | All operations logged to `~/.vmware-aiops/audit.log` (JSONL) with before/after state |\n| **Input Validation** | VM name, CPU (1-128), memory (128-1048576 MB), disk (1-65536 GB) validated |\n| **Password Protection** | `.env` file loading with permission check; never in shell history |\n| **SSL Self-signed Support** | `verify_ssl: false` — only for ESXi with self-signed certs in isolated labs; production should use CA-signed certificates |\n| **Prompt Injection Protection** | vSphere event messages and host logs are truncated, stripped of control characters, and wrapped in boundary markers before output |\n| **Webhook Data Scope** | Sends notifications to user-configured URLs only — no third-party services by default |\n| **Task Waiting** | All async operations wait for completion and report result |\n| **State Validation** | Pre-operation checks (VM exists, power state correct) |\n\n### vCenter vs ESXi Comparison\n\n| Capability | vCenter | ESXi Standalone |\n|------------|:-------:|:----:|\n| vMotion migration | ✅ | ❌ |\n| Cross-host clone | ✅ | ❌ |\n| Cluster management | ✅ | ❌ |\n| All VM lifecycle ops | ✅ | ✅ |\n| OVA/Template/Linked Clone deploy | ✅ | ✅ |\n| Datastore browsing & image scan | ✅ | ✅ |\n| Snapshots | ✅ | ✅ |\n| Guest operations | ✅ | ✅ |\n\n> Inventory, alarms, events, sensors, host services, and scanning are now in [vmware-monitor](https://github.com/vmware-skills/VMware-Monitor).\n\n### What protects you\n\nThe table above lists two different surfaces and it is worth being blunt about\nwhich protections apply to which, because getting this wrong is worse than\nhaving no protection at all — a guardrail you believe in is one you stop\ncompensating for.\n\n**On the CLI**, a destructive command asks twice and takes no bypass flag, and\n`--dry-run` previews any write. That defends a mistyped command typed by a\nhuman. It does not defend against an agent, which satisfies both prompts with\n`yes |`.\n\n**Over MCP**, there is no confirmation step at all. All 43 write tools —\n`vm_delete`, `cluster_delete`, `vm_guest_exec` among them — act on the first\ncall. Seven host-networking and DRS tools take a `confirm` argument that\ndefaults to a no-write preview, but that is a preview switch, not an approval\ngate: one more call is all it takes. This is deliberate. A `confirmed=`\nhandshake was considered in July 2026 and cut, along with the earlier\n`VMWARE_READ_ONLY` switch, because neither was a real boundary — the switch was\nenforced on the MCP path only and any agent with a shell walked around it via\nthe CLI, and a handshake is a speed-bump a model that intends to act steps over.\n\n**What actually decides whether a write lands is the vCenter/ESXi service\naccount.** Give the skill an account with the privileges the work needs and no\nmore; vCenter refuses the rest itself, on every surface, with no way around it\nfrom inside this skill. **To run an agent read-only, give it a read-only vCenter\nrole** — one decision, enforced where it is made. Every call is then recorded in\n`~/.vmware/audit.db` before the caller sees a result, which is how you find out\nwhat happened.\n\n**`vm_guest_exec` is the one to think hardest about.** It runs a caller-supplied\ncommand inside the guest OS with the credentials handed to it, which the\ndocumentation's own example makes `root`; nothing bounds what the command may\nbe. The guest account is a *separate* authorization boundary from the vCenter\none — a read-only vCenter role does not constrain what this tool does inside a\nVM. If you do not need guest operations, do not configure guest credentials.\n\nThe full inventory of which tools are gated and which are not is in\n[references/capabilities.md](skills/vmware-aiops/references/capabilities.md#what-gates-a-write),\nwhere the numbers are checked against the live tool registry by the test suite\nrather than maintained by hand.\n\n---\n\n## Troubleshooting\n\n### \"VM not found\" error\nVM names are case-sensitive in vSphere. Use exact name from `vmware-monitor inventory vms`.\n\n### Guest exec returns empty output\nUse `vm_guest_exec_output` instead of `vm_guest_exec` — it auto-captures stdout/stderr. Basic `vm_guest_exec` only returns exit code.\n\n### Deploy OVA times out\nLarge OVA files (>10GB) may exceed the default 120s timeout. The upload happens via HTTP NFC lease — ensure network between the machine running vmware-aiops and ESXi is stable.\n\n### Plan apply fails mid-way\nRun `vmware-aiops plan list` to see failed plan status. Ask user if they want to rollback with `vm_rollback_plan`. Irreversible steps (delete_vm) are skipped during rollback.\n\n### Connection refused / SSL error\n1. Verify target is reachable: `vmware-aiops doctor`\n2. For self-signed certs: set `verify_ssl: false` in config.yaml (lab environments only)\n\n---\n\n## Supported AI Platforms\n\n| Platform | Status | Config File | AI Model |\n|----------|--------|-------------|----------|\n| **Claude Code** | ✅ Native Skill | `skills/vmware-aiops/SKILL.md` | Anthropic Claude |\n| **Gemini CLI** | ✅ Context file + MCP | `skills/vmware-aiops/SKILL.md` | Google Gemini |\n| **OpenAI Codex CLI** | ✅ Skill + AGENTS.md | `skills/vmware-aiops/SKILL.md` | OpenAI GPT |\n| **Aider** | ✅ Conventions | `skills/vmware-aiops/SKILL.md` | Any (cloud + local) |\n| **Continue CLI** | ✅ Rules | `skills/vmware-aiops/SKILL.md` | Any (cloud + local) |\n| **Trae IDE** | ✅ Rules | `skills/vmware-aiops/SKILL.md` | Claude/DeepSeek/GPT-4o/Doubao |\n| **Kimi Code CLI** | ✅ Skill | `skills/vmware-aiops/SKILL.md` | Moonshot Kimi |\n| **MCP Server** | ✅ MCP Protocol | `vmware_aiops/mcp_server/` | Any MCP client |\n| **Python CLI** | ✅ Standalone | N/A | N/A |\n\n### Platform Comparison\n\n| Feature | Claude Code | Gemini CLI | Codex CLI | Aider | Continue | Trae IDE | Kimi CLI |\n|---------|-------------|------------|-----------|-------|----------|----------|----------|\n| Cloud AI | Anthropic | Google | OpenAI | Any | Any | Multi | Moonshot |\n| Local models | — | — | — | Ollama | Ollama | — | — |\n| Skill system | SKILL.md | Context file | SKILL.md | — | Rules | Rules | SKILL.md |\n| MCP support | Native | Native | Via Skills | Third-party | Native | — | — |\n| Free tier | — | 60 req/min | — | Self-hosted | Self-hosted | — | — |\n\n### MCP Server Integrations\n\nThe vmware-aiops MCP server works with **any MCP-compatible agent or tool**. Ready-to-use configuration templates are in [`examples/mcp-configs/`](examples/mcp-configs/).\n\n| Agent / Tool | Local Model Support | Config Template | Integration Guide |\n|-------------|:-------------------:|-----------------|-------------------|\n| **[Xiaoguai (小怪)](https://github.com/xiaoguai-agent/xiaoguai)** | ✅ Self-hosted, any LLM | [MCP setup](https://github.com/xiaoguai-agent/xiaoguai/blob/main/docs/book/src/api/mcp.md) | [Guide](https://github.com/xiaoguai-agent/xiaoguai) |\n| **[Goose](https://github.com/block/goose)** | ✅ Ollama, LM Studio | [`goose.json`](examples/mcp-configs/goose.json) | [Guide](docs/integrations/goose.md) |\n| **[LocalCowork](https://github.com/Liquid4All/localcowork)** | ✅ Fully offline | [`localcowork.json`](examples/mcp-configs/localcowork.json) | [Guide](docs/integrations/localcowork.md) |\n| **[mcp-agent](https://github.com/lastmile-ai/mcp-agent)** | ✅ Ollama, vLLM | [`mcp-agent.yaml`](examples/mcp-configs/mcp-agent.yaml) | [Guide](docs/integrations/mcp-agent.md) |\n| **VS Code Copilot** | — | [`vscode-copilot.json`](examples/mcp-configs/vscode-copilot.json) | [Guide](docs/integrations/vscode-copilot.md) |\n| **[Cursor](https://www.cursor.com)** | — | [`cursor.json`](examples/mcp-configs/cursor.json) | [Guide](docs/integrations/cursor.md) |\n| **Continue** | ✅ Ollama | [`continue.yaml`](examples/mcp-configs/continue.yaml) | [Guide](docs/integrations/continue.md) |\n| **Claude Code** | — | [`claude-code.json`](examples/mcp-configs/claude-code.json) | — |\n\n> **[Xiaoguai (小怪)](https://github.com/xiaoguai-agent/xiaoguai)** — a self-hostable, audit-first agent platform (Rust, single binary + embedded SQLite) from the same maintainer. It runs the vmware-aiops MCP server as one of its toolboxes; being both an MCP *consumer* and an MCP *server*, its HMAC-chained audit log and human-on-the-loop approval gates line up with this skill's own audit + confirm design. See its [MCP integration guide](https://github.com/xiaoguai-agent/xiaoguai/blob/main/docs/book/src/api/mcp.md).\n\n**Fully local operation** (no cloud API required):\n\n```bash\n# Aider + Ollama + vmware-aiops (via SKILL.md)\naider --conventions skills/vmware-aiops/SKILL.md --model ollama/qwen2.5-coder:32b\n\n# Any MCP agent + local model + vmware-aiops MCP server\n# See examples/mcp-configs/ for your agent's config format\n```\n\n---\n\n## Installation\n\n### Step 0: Prerequisites\n\n```bash\n# Python 3.10+ required\npython3 --version\n\n# Node.js 18+ required for Gemini CLI and Codex CLI\nnode --version\n```\n\n### Step 1: Clone & Install Python Backend\n\nAll platforms share the same Python backend.\n\n```bash\ngit clone https://github.com/vmware-skills/VMware-AIops.git\ncd VMware-AIops\npython3 -m venv .venv\nsource .venv/bin/activate\npip install -e .\n```\n\n### Step 2: Configure\n\n```bash\nmkdir -p ~/.vmware-aiops\ncp config.example.yaml ~/.vmware-aiops/config.yaml\n# Edit config.yaml with your vCenter/ESXi targets\n```\n\nSet passwords via `.env` file (recommended):\n```bash\n# Use the template\ncp .env.example ~/.vmware-aiops/.env\n\n# Edit and fill in your passwords, then lock permissions\nchmod 600 ~/.vmware-aiops/.env\n```\n\n> **Security note**: Prefer `.env` file over command-line `export` to avoid passwords appearing in shell history. The `.env` file should have `chmod 600` (owner-only read/write).\n\nPassword environment variable naming convention:\n```\nVMWARE_{TARGET_NAME_UPPER}_PASSWORD\n# Replace hyphens with underscores, UPPERCASE\n# Example: target \"home-esxi\" → VMWARE_HOME_ESXI_PASSWORD\n# Example: target \"prod-vcenter\" → VMWARE_PROD_VCENTER_PASSWORD\n```\n\n### Security Best Practices\n\n- **NEVER** hardcode passwords in scripts or config files\n- **NEVER** pass passwords as command-line arguments (visible in `ps`)\n- **ALWAYS** use `~/.vmware-aiops/.env` with `chmod 600`\n- **ALWAYS** configure connections via `config.yaml` — credentials are loaded from `.env` automatically\n- **Config File Contents**: `config.yaml` stores target hostnames, ports, and a reference to the `.env` file. It does **not** contain passwords or tokens. All secrets are stored exclusively in `.env`\n- **TLS**: Enabled by default. Disable only for ESXi hosts with self-signed certificates in isolated lab environments\n- **Webhook**: Disabled by default. When enabled, sends monitoring summaries to your own configured URL only — payloads contain no credentials, IPs, or PII, only aggregated alert metadata. No data sent to third-party services\n- **Least Privilege**: Use a dedicated vCenter service account with minimal permissions. For monitoring-only use cases, prefer the read-only [VMware-Monitor](https://github.com/vmware-skills/VMware-Monitor)\n- **Prompt Injection Protection**: All vSphere-sourced content is truncated, stripped of control characters, and wrapped in boundary markers before output\n- **Code Review**: We recommend reviewing the [source code](https://github.com/vmware-skills/VMware-AIops) and commit history before deploying in production\n- **Production Safety**: For production environments, use the read-only [VMware-Monitor](https://github.com/vmware-skills/VMware-Monitor) instead. AI agents can misinterpret context and execute unintended destructive operations — real-world incidents have shown that AI-driven infrastructure tools without proper isolation can delete production databases and entire environments. VMware-Monitor eliminates this risk at the code level: no destructive functions exist in its codebase\n\n### Step 3: Connect Your AI Tool\n\nChoose one (or more) of the following:\n\n---\n\n#### Option A: Claude Code\n\n**Method 1: Skills.sh or ClawHub (recommended)**\n\nEither installer places the skill in Claude Code's skills directory for you:\n\n```bash\nnpx skills add vmware-skills/VMware-AIops\n# or\nclawhub install @zw008/vmware-aiops\n```\n\n**Method 2: Manual skill install**\n\n```bash\ngit clone https://github.com/vmware-skills/VMware-AIops.git\ncd VMware-AIops\n\n# Copy the skill into Claude Code's personal skills directory\nmkdir -p ~/.claude/skills/vmware-aiops\ncp -r skills/vmware-aiops/. ~/.claude/skills/vmware-aiops/\n```\n\nFor tool access (not just skill context), also register the MCP server:\n\n```bash\nclaude mcp add vmware-aiops -- vmware-aiops mcp\n```\n\nRestart Claude Code, then:\n```\n> Show me all VMs on esxi-lab.example.com\n```\n\n**Submit to Official Marketplace**\n\nThis plugin can also be submitted to the [Anthropic official plugin directory](https://clau.de/plugin-directory-submission) for public discovery.\n\n---\n\n#### Option B: Gemini CLI\n\n```bash\n# Install Gemini CLI\nnpm install -g @google/gemini-cli\n\n# Load the skill as project context (Gemini CLI reads GEMINI.md on startup)\ncp skills/vmware-aiops/SKILL.md ./GEMINI.md\n```\n\nFor tool access (not just context), register the MCP server in `~/.gemini/settings.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"vmware-aiops\": {\n      \"command\": \"vmware-aiops\",\n      \"args\": [\"mcp\"],\n      \"env\": { \"VMWARE_AIOPS_CONFIG\": \"~/.vmware-aiops/config.yaml\" }\n    }\n  }\n}\n```\n\nThen start Gemini CLI:\n```\ngemini\n> Show me all VMs on my ESXi host\n```\n\n---\n\n#### Option C: OpenAI Codex CLI\n\n```bash\n# Install Codex CLI\nnpm i -g @openai/codex\n# Or on macOS:\n# brew install --cask codex\n\n# Copy skill to Codex skills directory\nmkdir -p ~/.codex/skills/vmware-aiops\ncp skills/vmware-aiops/SKILL.md ~/.codex/skills/vmware-aiops/SKILL.md\n\n# Copy AGENTS.md to project root\ncp skills/vmware-aiops/SKILL.md ./AGENTS.md\n```\n\nThen start Codex CLI:\n```bash\ncodex --enable skills\n> List all VMs on my ESXi\n```\n\n---\n\n#### Option D: Aider (supports local models)\n\n```bash\n# Install Aider\npip install aider-chat\n\n# Install Ollama for local models (optional)\n# macOS:\nbrew install ollama\nollama pull qwen2.5-coder:32b\n\n# Run with cloud API\naider --conventions skills/vmware-aiops/SKILL.md\n\n# Or with local model via Ollama\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model ollama/qwen2.5-coder:32b\n```\n\n---\n\n#### Option E: Continue CLI (supports local models)\n\n```bash\n# Install Continue CLI\nnpm i -g @continuedev/cli\n\n# Copy rules file\nmkdir -p .continue/rules\ncp skills/vmware-aiops/SKILL.md .continue/rules/vmware-aiops.md\n```\n\nConfigure `~/.continue/config.yaml` for local model:\n```yaml\nmodels:\n  - name: local-coder\n    provider: ollama\n    model: qwen2.5-coder:32b\n```\n\nThen:\n```bash\ncn\n> Check ESXi health and alarms\n```\n\n---\n\n#### Option F: Trae IDE\n\nCopy the rules file to your project's `.trae/rules/` directory:\n\n```bash\nmkdir -p .trae/rules\ncp skills/vmware-aiops/SKILL.md .trae/rules/project_rules.md\n```\n\nTrae IDE's Builder Mode reads `.trae/rules/` Markdown files at startup.\n\n> **Note**: You can also install Claude Code extension in Trae IDE and use `.claude/skills/` format directly.\n\n---\n\n#### Option G: Kimi Code CLI\n\n```bash\n# Copy skill file to Kimi skills directory\nmkdir -p ~/.kimi/skills/vmware-aiops\ncp skills/vmware-aiops/SKILL.md ~/.kimi/skills/vmware-aiops/SKILL.md\n```\n\n---\n\n#### Option H: MCP Server (Glama / Claude Desktop)\n\nThe MCP server exposes VMware operations as tools via the [Model Context Protocol](https://modelcontextprotocol.io). Works with any MCP-compatible client (Claude Desktop, Cursor, etc.).\n\n**After `uv tool install vmware-aiops`, start the MCP server with one command** (v1.5.15+):\n\n```bash\n# Recommended — single command, no network re-resolve\nvmware-aiops mcp\n\n# With a custom config path\nVMWARE_AIOPS_CONFIG=/path/to/config.yaml vmware-aiops mcp\n```\n\n**Claude Desktop config** (`claude_desktop_config.json`):\n```json\n{\n  \"mcpServers\": {\n    \"vmware-aiops\": {\n      \"command\": \"vmware-aiops\",\n      \"args\": [\"mcp\"],\n      \"env\": {\n        \"VMWARE_AIOPS_CONFIG\": \"/path/to/config.yaml\"\n      }\n    }\n  }\n}\n```\n\n<details>\n<summary>Alternative: uvx (no install) or legacy entry point</summary>\n\n```bash\n# Run without installing (requires PyPI access each launch)\nuvx --from vmware-aiops vmware-aiops mcp\n\n# Legacy entry point (still works, kept for backward compatibility)\nvmware-aiops-mcp\n```\n\n> **Behind a corporate TLS proxy?** uvx may fail with `invalid peer certificate: UnknownIssuer`.\n> Use the recommended `vmware-aiops mcp` form above (no network needed), or set `UV_NATIVE_TLS=true`.\n\n</details>\n\n\n---\n\n#### Option I: Standalone CLI (no AI)\n\n```bash\n# Already installed in Step 1\nsource .venv/bin/activate\n\nvmware-aiops vm power-on my-vm --target home-esxi\nvmware-aiops deploy ova ./ubuntu.ova --name my-vm --target home-esxi\nvmware-aiops datastore browse datastore1 --target home-esxi\n```\n\n---\n\n## Update / Upgrade\n\nAlready installed? Re-run the install command for your channel to get the latest version:\n\n| Install Channel | Update Command |\n|----------------|----------------|\n| ClawHub | `clawhub install @zw008/vmware-aiops` |\n| Skills.sh | `npx skills add vmware-skills/VMware-AIops` |\n| Git clone | `cd VMware-AIops && git pull origin main && uv pip install -e .` |\n| uv | `uv tool install vmware-aiops --force` |\n\nCheck your current version: `vmware-aiops --version`\n\n---\n\n## Chinese Cloud Models\n\nFor users in China who prefer domestic cloud APIs or have limited access to overseas services.\n\n### DeepSeek\n\nCost-effective, strong coding capability.\n\n```bash\n# Set DeepSeek API key (get from https://platform.deepseek.com)\nexport DEEPSEEK_API_KEY=\"your-key\"\n\n# Run with Aider\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model deepseek/deepseek-coder\n```\n\nPersistent config `~/.aider.conf.yml`:\n```yaml\nmodel: deepseek/deepseek-coder\nconventions: skills/vmware-aiops/SKILL.md\n```\n\n### Qwen (Alibaba Cloud)\n\nAlibaba Cloud's coding model, free tier available.\n\n```bash\n# Set DashScope API key (get from https://dashscope.console.aliyun.com)\nexport DASHSCOPE_API_KEY=\"your-key\"\n\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model qwen/qwen-coder-plus\n```\n\nOr via OpenAI-compatible endpoint:\n```bash\nexport OPENAI_API_BASE=\"https://dashscope.aliyuncs.com/compatible-mode/v1\"\nexport OPENAI_API_KEY=\"your-dashscope-key\"\n\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model qwen-coder-plus-latest\n```\n\n### Doubao (ByteDance)\n\n```bash\nexport OPENAI_API_BASE=\"https://ark.cn-beijing.volces.com/api/v3\"\nexport OPENAI_API_KEY=\"your-ark-key\"\n\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model your-doubao-endpoint-id\n```\n\n### With Continue CLI\n\nConfigure `~/.continue/config.yaml`:\n\n```yaml\n# DeepSeek\nmodels:\n  - name: deepseek-coder\n    provider: openai-compatible\n    apiBase: https://api.deepseek.com/v1\n    apiKey: your-deepseek-key\n    model: deepseek-coder\n\n# Qwen\nmodels:\n  - name: qwen-coder\n    provider: openai-compatible\n    apiBase: https://dashscope.aliyuncs.com/compatible-mode/v1\n    apiKey: your-dashscope-key\n    model: qwen-coder-plus-latest\n```\n\n---\n\n## Local Models (Aider + Ollama)\n\nFor fully offline operation — no cloud API, no internet, full privacy.\n\n**Aider + Ollama + local Qwen/DeepSeek** is ideal for air-gapped environments.\n\n### Step 1: Install Ollama\n\n```bash\n# macOS\nbrew install ollama\n\n# Linux — download from https://ollama.com/download and install manually\n# See https://github.com/ollama/ollama for platform-specific instructions\n```\n\n### Step 2: Pull a model\n\n| Model | Command | Size | Note |\n|-------|---------|------|------|\n| **Qwen 2.5 Coder 32B** | `ollama pull qwen2.5-coder:32b` | ~20GB | Best local coding model |\n| **Qwen 2.5 Coder 7B** | `ollama pull qwen2.5-coder:7b` | ~4.5GB | Low-memory option |\n| **DeepSeek Coder V2** | `ollama pull deepseek-coder-v2` | ~8.9GB | Strong reasoning |\n| **CodeLlama 34B** | `ollama pull codellama:34b` | ~19GB | Meta coding model |\n\n> **Hardware**: 32B → ~20GB VRAM (or 32GB RAM for CPU). 7B → 8GB RAM.\n\n### Step 3: Run with Aider\n\n```bash\npip install aider-chat\nollama serve\n\n# Aider + local Qwen (recommended)\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model ollama/qwen2.5-coder:32b\n\n# Aider + local DeepSeek\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model ollama/deepseek-coder-v2\n\n# Low-memory option\naider --conventions skills/vmware-aiops/SKILL.md \\\n  --model ollama/qwen2.5-coder:7b\n```\n\nPersistent config `~/.aider.conf.yml`:\n```yaml\nmodel: ollama/qwen2.5-coder:32b\nconventions: skills/vmware-aiops/SKILL.md\n```\n\n### Local Architecture\n\n```\nUser → Aider CLI → Ollama (localhost:11434) → Qwen / DeepSeek local model\n  │                                                    ↓\n  │                                          reads AGENTS.md instructions\n  │                                                    ↓\n  └──────────────────────────────→ vmware-aiops CLI ──→ ESXi / vCenter\n```\n\n> **Tip**: Local models are fully offline — perfect for air-gapped environments or strict data compliance.\n\n---\n\n## CLI Reference\n\n```bash\n# Diagnostics\nvmware-aiops doctor                   # Check environment, config, connectivity\nvmware-aiops doctor --skip-auth       # Skip vSphere auth check (faster)\n\n# MCP Config Generator\nvmware-aiops mcp-config generate --agent goose        # Generate config for Goose\nvmware-aiops mcp-config generate --agent claude-code  # Generate config for Claude Code\nvmware-aiops mcp-config list                          # List all supported agents\n\n# VM operations\nvmware-aiops vm power-on my-vm                                 # Power on\nvmware-aiops vm power-off my-vm                                # Graceful shutdown (2x confirm)\nvmware-aiops vm power-off my-vm --force                        # Force power off (2x confirm)\nvmware-aiops vm create my-new-vm --cpu 4 --memory 8192 --disk 100  # Create VM\nvmware-aiops vm delete my-vm --confirm                         # Delete VM (2x confirm)\nvmware-aiops vm reconfigure my-vm --cpu 4 --memory 8192        # Reconfigure (2x confirm)\nvmware-aiops vm snapshot-create my-vm --name \"before-upgrade\"  # Create snapshot\nvmware-aiops vm snapshot-list my-vm                            # List snapshots\nvmware-aiops vm snapshot-revert my-vm --name \"before-upgrade\"  # Revert snapshot\nvmware-aiops vm snapshot-delete my-vm --name \"before-upgrade\"  # Delete snapshot (waits ≤30 min for consolidation)\nvmware-aiops vm snapshot-delete my-vm --name \"old-big\" --no-wait  # Fire async, return a task id\nvmware-aiops vm task-status task-1234                          # Poll an async task by id\nvmware-aiops vm clone my-vm --new-name my-vm-clone             # Clone VM\nvmware-aiops vm migrate my-vm --to-host esxi-02                # vMotion\nvmware-aiops vm set-ttl my-vm --minutes 60                     # Auto-delete in 60 min\nvmware-aiops vm cancel-ttl my-vm                               # Cancel TTL\nvmware-aiops vm list-ttl                                       # Show all TTLs\nvmware-aiops vm clean-slate my-vm --snapshot baseline          # Revert to baseline (2x confirm)\n\n# Guest Operations (requires VMware Tools in guest)\nvmware-aiops vm guest-exec my-vm --cmd /bin/bash --args \"-c 'whoami'\" --user root\nvmware-aiops vm guest-upload my-vm --local ./script.sh --guest /tmp/script.sh --user root\nvmware-aiops vm guest-download my-vm --guest /var/log/syslog --local ./syslog.txt --user root\n\n# Plan → Apply (multi-step operations)\nvmware-aiops plan list                                        # List pending/failed plans\n\n# Deploy\nvmware-aiops deploy ova ./ubuntu.ova --name my-vm --datastore ds1      # Deploy from OVA\nvmware-aiops deploy template golden-ubuntu --name new-vm               # Deploy from template\nvmware-aiops deploy linked-clone --source base-vm --snapshot clean --name test-vm  # Linked clone (seconds)\nvmware-aiops deploy iso my-vm --iso \"[datastore1] iso/ubuntu-22.04.iso\"  # Attach ISO\nvmware-aiops deploy mark-template golden-vm                            # Convert VM to template\nvmware-aiops deploy batch-clone --source base-vm --count 5 --prefix lab  # Batch clone\nvmware-aiops deploy batch deploy.yaml                                  # Batch deploy from YAML spec\n\n# Cluster\nvmware-aiops cluster info my-cluster                                   # Cluster details (HA/DRS status)\nvmware-aiops cluster create my-cluster --ha --drs                      # Create cluster with HA+DRS\nvmware-aiops cluster delete my-cluster                                 # Delete cluster (2x confirm)\nvmware-aiops cluster add-host my-cluster --host esxi-03                # Add host to cluster (2x confirm)\nvmware-aiops cluster remove-host my-cluster --host esxi-03             # Remove host (2x confirm)\nvmware-aiops cluster configure my-cluster --ha --drs                   # Configure HA/DRS (2x confirm)\n\n# Alarm management\nvmware-aiops alarm list                                                # List triggered alarms\nvmware-aiops alarm acknowledge esxi-01 \"Host memory usage\"             # Acknowledge alarm\nvmware-aiops alarm reset esxi-01 \"Host memory usage\"                   # Clear alarms (2x confirm; clears ALL matching entity type + status)\n\n# Datastore (browse and scan only — iSCSI/vSAN moved to vmware-storage)\nvmware-aiops datastore browse datastore1 --path \"iso/\"                 # Browse datastore\nvmware-aiops datastore scan-images --target home-esxi                  # Scan all datastores for images\n\n# Scan\nvmware-aiops scan now              # One-time scan\n\n# Daemon\nvmware-aiops daemon start          # Start scanner\nvmware-aiops daemon status         # Check status\nvmware-aiops daemon stop           # Stop daemon\n\n# Companion skills for other operations:\n#   vmware-monitor: inventory, alarms, events, sensors\n#   vmware-storage: datastores, iSCSI, vSAN\n#   vmware-vks:     Tanzu/TKC cluster lifecycle\n```\n\n---\n\n## Configuration\n\nSee `config.example.yaml` for all options.\n\n| Section | Key | Default | Description |\n|---------|-----|---------|-------------|\n| targets | name | — | Friendly name |\n| targets | host | — | vCenter/ESXi hostname or IP |\n| targets | type | vcenter | `vcenter` or `esxi` |\n| targets | port | 443 | Connection port |\n| targets | verify_ssl | true | Verify the target's TLS certificate (set false only for self-signed lab hosts) |\n| scanner | interval_minutes | 15 | Scan frequency |\n| scanner | severity_threshold | warning | Min severity: critical/warning/info |\n| scanner | lookback_hours | 1 | How far back to scan |\n| scanner | log_types | [vpxd, hostd, vmkernel] | Log sources |\n| notify | log_file | ~/.vmware-aiops/scan.log | JSONL log output |\n| notify | webhook_url | — | Webhook endpoint (Slack, Discord, etc.) |\n\n---\n\n## Project Structure\n\n```\nVMware-AIops/\n├── skills/                        # Skills index (npx skills add)\n│   └── vmware-aiops/\n│       ├── SKILL.md               # Slimmed-down skill (progressive disclosure)\n│       └── references/            # Detailed docs loaded on-demand\n│           ├── capabilities.md    # Full capabilities tables\n│           ├── cli-reference.md   # Complete CLI reference\n│           └── setup-guide.md     # Install, security, AI platforms\n├── vmware_aiops/                  # Python backend\n│   ├── config.py                  # YAML + .env config\n│   ├── connection.py              # Multi-target pyVmomi\n│   ├── cli/                       # Typer CLI (double confirm)\n│   ├── ops/                       # Operations\n│   │   ├── inventory.py           # VMs, hosts, datastores, clusters\n│   │   ├── health.py              # Alarms, events, sensors\n│   │   ├── vm_lifecycle.py        # VM CRUD, snapshots, clone, migrate\n│   │   ├── vm_deploy.py           # OVA, template, linked clone, batch deploy\n│   │   └── datastore_browser.py   # Datastore browsing, image discovery\n│   ├── scanner/                   # Log scanning daemon\n│   ├── notify/                    # Notifications (JSONL + webhook)\n│   └── mcp_server/                # MCP server wrapper\n│       ├── server.py              # FastMCP server with tools\n│       └── __main__.py\n├── examples/mcp-configs/          # MCP client config templates\n├── tests/                         # Test suite\n├── smithery.yaml                  # Smithery marketplace config\n├── RELEASE_NOTES.md\n├── config.example.yaml\n└── pyproject.toml\n```\n\n## API Coverage\n\nBuilt on **pyVmomi** (vSphere Web Services API / SOAP).\n\n| API Object | Usage |\n|------------|-------|\n| `vim.VirtualMachine` | VM lifecycle, snapshots, clone, migrate |\n| `vim.HostSystem` | ESXi host info, sensors, services |\n| `vim.Datastore` | Storage capacity, type, accessibility |\n| `vim.host.DatastoreBrowser` | File browsing, image discovery (ISO/OVA/VMDK) |\n| `vim.OvfManager` | OVA import and deployment |\n| `vim.ClusterComputeResource` | Cluster, DRS, HA |\n| `vim.Network` | Network listing |\n| `vim.alarm.AlarmManager` | Active alarm monitoring |\n| `vim.event.EventManager` | Event/log queries |\n\n## Related Projects\n\n| Skill | Scope | Tools | Install |\n|-------|-------|:-----:|---------|\n| **[vmware-aiops](https://github.com/vmware-skills/VMware-AIops)** | VM lifecycle, deployment, guest ops, cluster, datastore browse, triage | 49 | `uv tool install vmware-aiops` |\n| **[vmware-monitor](https://github.com/vmware-skills/VMware-Monitor)** | Read-only monitoring, alarms, events, investigation bundles | 27 | `uv tool install vmware-monitor` |\n| **[vmware-storage](https://github.com/vmware-skills/VMware-Storage)** | Datastores, iSCSI, vSAN | 11 | `uv tool install vmware-storage` |\n| **[vmware-vks](https://github.com/vmware-skills/VMware-VKS)** | Tanzu Namespaces, TKC cluster lifecycle | 20 | `uv tool install vmware-vks` |\n| **[vmware-nsx](https://github.com/vmware-skills/VMware-NSX)** | NSX networking: segments, gateways, NAT, routing, IPAM | 33 | `uv tool install vmware-nsx-mgmt` |\n| **[vmware-nsx-security](https://github.com/vmware-skills/VMware-NSX-Security)** | DFW policies/rules, security groups, Traceflow, IDS/IPS | 21 | `uv tool install vmware-nsx-security` |\n| **[vmware-aria](https://github.com/vmware-skills/VMware-Aria)** | Aria Operations metrics, alerts, capacity, anomalies | 28 | `uv tool install vmware-aria` |\n| **[vmware-avi](https://github.com/vmware-skills/VMware-AVI)** | AVI (NSX ALB) load balancing, AKO Kubernetes ops | 28 | `uv tool install vmware-avi` |\n| **[vmware-harden](https://github.com/vmware-skills/VMware-Harden)** | Compliance baselines (CIS / vSphere SCG / 等保 / PCI-DSS), drift detection | 6 | `uv tool install vmware-harden` |\n\n---\n\n## Troubleshooting & Contributing\n\nIf you encounter any errors or issues, please send the error message, logs, or screenshots to **zhouwei008@gmail.com**. Contributions are welcome — feel free to join us in maintaining and improving this project!\n\n## License\n\nMIT\n",
  "bytes": 48529,
  "sha": "02782001d87a2c839bf030ed2a981156dc8eaa7e41746a4631659eb5175b5027",
  "repo_slug": "vmware-skills/vmware-aiops",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_vmware_skills_vmware_aiops_07e67c07/readme"
}