5,183,178 AI resources indexedWorking toward the world’s largest AI resource index.

Proxmox Skill

Manage Proxmox VE nodes, VMs, and containers. Can list hardware stats, resources, and control power states (start, stop, reboot, shutdown).

Installable skillMITBy kbarbel640-delUpdated Oct 31, 2026
INTENDED FOROpenClawInstall in a listed agent.

instructions.md

---
name: proxmox-06429424b0c9
description: "Manage Proxmox VE nodes, VMs, and containers. Can list hardware stats, resources, and control power states (start, stop, reboot, shutdown)."
metadata:
  requires:
    bins: ["python3"]
  env: ["PVE_HOST", "PVE_TOKEN_ID", "PVE_TOKEN_SECRET"]
---

# Proxmox Skill

This skill allows the agent to interact with a Proxmox VE cluster to manage virtual machines and containers.

## Tools

### proxmox_list
List Proxmox nodes or all available VMs and containers across the entire cluster.
- Command: python3 {{skillDir}}/scripts/proxmox.py {{type}}
- Args:
  - type: "nodes" or "vms"

### proxmox_node_health
Get hardware-level health stats (CPU usage, RAM, Uptime, Version) for a specific physical node.
- Command: python3 {{skillDir}}/scripts/proxmox.py node_health {{node}}
- Args:
  - node: The name of the Proxmox host (e.g., "pve" or "hydra")

### proxmox_status
Get the real-time status of a specific VM or container.
- Command: python3 {{skillDir}}/scripts/proxmox.py status {{node}} {{kind}} {{vmid}}
- Args:
  - node: The Proxmox node name where the resource lives
  - kind: "qemu" for VMs, "lxc" for containers
  - vmid: The numerical ID of the resource (e.g., "100")

### proxmox_power_action
Perform power management actions. These actions require human approval by default.
- Approval: true
- Command: python3 {{skillDir}}/scripts/proxmox.py {{action}} {{node}} {{kind}} {{vmid}}
- Args:
  - action: "start", "stop", "reboot", or "shutdown"
  - node: The Proxmox node name
  - kind: "qemu" or "lxc"
  - vmid: The ID of the resource

Discuss this resource

Platform guidance

Imported from the listed AgentSkill.sh category. Complete supplied package files and notices are retained. Includes source scripts; review them before use. Required external tools and credentials are not installed. No bundled code was executed during import. Individual agent task behavior has not been tested.