Use when managing Proxmox VMs, LXC containers, storage, snapshots, or running commands inside VMs. Uses Proxmox MCP tools directly.
Host: 100.99.135.123:8006 | Token: root@pam!mcpserver
| ID | Name | Type | IP | SSH | Notes |
|---|---|---|---|---|---|
| 100 | Windows11-Gaming | QEMU | — | — | Stopped, gaming rig |
| 101 | CasaOS | QEMU | 192.168.50.19 | nerfherder@ pubkey | Docker host, all services |
| 102 | recall-ollama | QEMU | 192.168.50.62 / Tailscale 100.70.195.84 | nerfherder / lacetimcat1216 | GPU inference, Recall API |
| 103 | ubuntu-claude | QEMU | 192.168.50.35 | nerfherder@ pubkey (no password) | Claude Code runner, QEMU agent installed |
| 104 |
| bricklayer-train |
| LXC |
| — |
| — |
| BrickLayer training |
| 105 | codevvos | LXC | — | — | CodevvOS project |
proxmox_get_vms → list all VMs with status
proxmox_get_vm_status → node=pve, vmid=103
proxmox_get_node_status → node=pve
proxmox_get_storage → node=pve
proxmox_start_vm → node=pve, vmid=103
proxmox_stop_vm → node=pve, vmid=103 (hard stop)
proxmox_reboot_vm → node=pve, vmid=103 (soft reboot — does NOT add new hardware)
Note: Config changes (agent, hardware) require full stop+start, not just reboot.
proxmox_execute_vm_command → node=pve, vmid=103, command="df -h /"
proxmox_resize_disk_vm → node=pve, vmid=103, disk=scsi0, size=+20G
sudo growpart /dev/sda 3
sudo pvresize /dev/sda3
sudo lvextend -l +100%FREE /dev/ubuntu-vg/ubuntu-lv
sudo resize2fs /dev/ubuntu-vg/ubuntu-lv
proxmox_create_snapshot_vm → node=pve, vmid=103, snapname="before-change"
proxmox_list_snapshots_vm → node=pve, vmid=103
proxmox_rollback_snapshot_vm → node=pve, vmid=103, snapname="before-change"
curl -s -k -X PUT https://100.99.135.123:8006/api2/json/nodes/pve/qemu/VMID/config \
-H "Authorization: PVEAPIToken=root@pam!mcpserver=69c9ecf2-c642-4017-8712-0654fa799c9a" \
-d "agent=1"
sudo apt install -y qemu-guest-agent && sudo systemctl enable --now qemu-guest-agentlocal — ISOs, backups, templates (93GB, 18% used)local-lvm — VM disks, LXC rootfs (1.71TB, ~50% used)# Disk usage
df -h /
# Large files
find / -xdev -size +200M
# Docker cleanup
docker system prune -f
# Journal logs
journalctl --vacuum-size=200M
# Apt cache
apt clean