proxmox-mcp-server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LOG_LEVEL | No | Log level: debug, info, warn, error (default info) | info |
| PROXMOX_HOST | Yes | Full base URL of the Proxmox server, e.g. https://pve.local:8006 | |
| PVE_READONLY | No | Set to 'true' (default) to register only read-only tools; 'false' enables write tools | true |
| PROXMOX_TOKEN_ID | Yes | API token ID in the format USER@REALM!TOKENID | |
| PVE_NODE_ALLOWLIST | No | Optional comma-separated list of node names to restrict operations | |
| PVE_VMID_ALLOWLIST | No | Optional comma-separated list of guest IDs to restrict operations | |
| PVE_TASK_TIMEOUT_MS | No | Timeout for waiting for asynchronous tasks in milliseconds (default 600000) | 600000 |
| PROXMOX_INSECURE_TLS | No | Set to 'true' to accept self-signed certificates (only for labs) | false |
| PROXMOX_TOKEN_SECRET | Yes | API token secret (UUID) | |
| PROXMOX_REQUEST_TIMEOUT_MS | No | Per-request timeout in milliseconds (default 30000) | 30000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| pve_versionA | Proxmox VE version of the connected node (version, release, repository id). Tells you which version-gated features are available. |
| pve_list_nodesA | List all nodes in the Proxmox cluster with their online status and load. |
| pve_node_statusA | Detailed status of a single node: CPU, memory, uptime, kernel and PVE version. |
| pve_cluster_resourcesA | Single combined view of cluster resources (VMs, containers, storage, nodes). Use the type filter to narrow the result. |
| pve_list_tasksA | Recent task log for a node, including UPID, type, status and exit status. Useful to check on a long-running operation that may have outlived a tool call timeout. |
| pve_list_vmsA | List QEMU VMs. With a node, lists that node directly; without one, returns every VM in the cluster from the resource index. |
| pve_vm_statusA | Current runtime status of a VM (running/stopped, CPU, memory, uptime). |
| pve_vm_configA | Full configuration of a VM (cores, memory, disks, network, boot order). |
| pve_list_containersA | List LXC containers. With a node, lists that node directly; without one, returns every container in the cluster from the resource index. |
| pve_lxc_statusA | Current runtime status of a container (running/stopped, CPU, memory, uptime). |
| pve_lxc_configA | Full configuration of a container (cores, memory, rootfs, network, features). |
| pve_list_snapshotsA | List snapshots of a VM or container, including the "current" pseudo-snapshot. |
| pve_list_storageA | List storages visible from a node, with type, enabled content and free space. |
| pve_storage_contentA | List the volumes on a storage: ISO images, container templates, disk images or backups. Use the content filter to narrow the listing. |
| pve_list_networkA | List network interfaces on a node (bridges, bonds, physical NICs, VLANs). Pending, not-yet-applied changes are included. |
| pve_list_backupsB | List backup archives stored on a storage. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 16 tools
Each tool targets a distinct resource and action, e.g., pve_list_vms vs pve_list_containers vs pve_list_backups. Even the combined view pve_cluster_resources is differentiated by its aggregate nature and type filter, so no ambiguity.
All tools follow a consistent pattern: 'pve_' prefix + verb_noun (e.g., pve_list_vms) or resource_config/status (e.g., pve_vm_config). The pattern is uniform and predictable across all 16 tools.
16 tools cover the major Proxmox domains: VMs, containers, storage, network, nodes, backups, snapshots, tasks, version, and cluster resources. This is a well-scoped set for a read-only monitoring server—neither too few nor too many.
The tool surface covers all key Proxmox resources for querying state. Minor gaps include no tool for individual node storage details beyond listing, and no mutation tools, but that is likely intentional given the server's apparent read-only purpose.