Proxmox VE MCP Server

Read from the package source without running it. The installed server may list more.

• pve_bulk_migrate_guests

  Migrate multiple VMs/containers to a target node in one bulk action (PVE 9.1+). Returns a task ID

• pve_bulk_shutdown_guests

  Gracefully shut down multiple VMs/containers cluster-wide in one bulk action (PVE 9.1+). Returns a task ID

• pve_bulk_start_guests

  Start multiple VMs/containers cluster-wide in one bulk action (PVE 9.1+). Returns a task ID

• pve_bulk_suspend_guests

  Suspend multiple VMs cluster-wide in one bulk action (PVE 9.1+). Returns a task ID

• pve_clone_lxc_container

  Clone an LXC container to create a new container from it

• pve_clone_qemu_vm

  Clone a QEMU VM to create a new VM from it

• pve_create_backup_jobwrite action

  Create a new scheduled backup job

• pve_create_firewall_rulewrite action

  Create a new cluster-level firewall rule

• pve_create_ha_resourcewrite action

  Add a VM or container to HA management. SID format: type:vmid (e.g. vm:100). Requires Sys.Console privilege on PVE 9.2+

• pve_create_ha_rulewrite action

  Create an HA rule (node-affinity: pin resources to nodes; resource-affinity: keep resources together or apart). Requires Sys.Console. Resources format: comma-separated IDs (e.g. vm:100,ct:200)

• pve_create_lxc_containerwrite action

  Create a new LXC container

• pve_create_lxc_snapshotwrite action

  Create a snapshot of an LXC container

• pve_create_networkwrite action

  Create a new network interface on a node

• pve_create_poolwrite action

  Create a new resource pool

• pve_create_qemu_snapshotwrite action

  Create a snapshot of a QEMU VM

• pve_create_qemu_vmwrite action

  Create a new QEMU virtual machine

• pve_create_userwrite action

  Create a new user in the PVE access control system

• pve_create_user_tokenwrite action

  Create an API token for a user. The token secret is returned ONCE in the response and cannot be retrieved again

• pve_delete_backup_jobwrite action

  Delete a scheduled backup job

• pve_delete_firewall_rulewrite action

  Delete a cluster-level firewall rule by position

• pve_delete_ha_resourcewrite action

  Remove a VM or container from HA management (does not delete the VM/container itself)

• pve_delete_ha_rulewrite action

  Delete an HA rule. Requires Sys.Console

• pve_delete_lxc_containerwrite action

  Delete an LXC container and all its data. The container must be stopped first.

• pve_delete_lxc_snapshotwrite action

  Delete a snapshot of an LXC container

• pve_delete_networkwrite action

  Delete a network interface configuration on a node

• pve_delete_poolwrite action

  Delete a resource pool (pool must be empty)

• pve_delete_qemu_snapshotwrite action

  Delete a snapshot of a QEMU VM

• pve_delete_qemu_vmwrite action

  Delete a QEMU virtual machine and all its data. The VM must be stopped first.

• pve_delete_userwrite action

  Delete a user from the PVE access control system

• pve_delete_user_tokenwrite action

  Delete an API token from a user

• pve_disarm_ha

  Disarm the HA stack cluster-wide for maintenance (releases node watchdogs, PVE 9.2+). resource_mode 'freeze': queue state changes until re-armed; 'ignore': HA ignores resources entirely. Requires Sys.Console privilege — a privilege-separated API token must have Sys.Console granted

• pve_get_backup_job

  Get the configuration of a specific backup job

• pve_get_cluster_ha_status

  Get the current HA manager status

• pve_get_cluster_log

  Get recent cluster log entries

• pve_get_cluster_options

  Get cluster-wide datacenter options

• pve_get_cluster_status

  Get the current cluster status including node membership and quorum

• pve_get_firewall_options

  Get the cluster-level firewall options

• pve_get_ha_resource

  Get the HA configuration for a specific resource. SID format: type:vmid (e.g. vm:100)

• pve_get_ha_rule

  Get the configuration of a specific HA rule

• pve_get_lxc_config

  Get the configuration of an LXC container

• pve_get_lxc_rrddata

  Get RRD statistics (CPU, memory, disk, network) for an LXC container over a time period

• pve_get_lxc_status

  Get the current status of an LXC container including CPU, memory, and disk usage

• pve_get_network

  Get the configuration of a specific network interface on a node

• pve_get_next_vmid

  Get the next available VMID in the cluster

• pve_get_node_dns

  Get DNS settings for a specific node

• pve_get_node_status

  Get detailed status of a specific node including CPU, memory, uptime, and load

• pve_get_node_syslog

  Get system log entries from a specific node

• pve_get_node_time

  Get time and timezone information for a specific node

• pve_get_node_version

  Get the PVE version information for a specific node

• pve_get_pool

  Get detailed information about a resource pool including its members

• pve_get_qemu_config

  Get the configuration of a QEMU VM

• pve_get_qemu_rrddata

  Get RRD statistics (CPU, memory, disk, network) for a QEMU VM over a time period

• pve_get_qemu_status

  Get the current status of a QEMU VM including CPU, memory, disk, and network usage

• pve_get_storage_config

  Get the configuration of a specific storage backend

• pve_get_storage_status

  Get the status and usage of a specific storage on a node

• pve_get_user

  Get detailed information about a specific user

• pve_list_acls

  List all access control list entries

• pve_list_backup_jobs

  List all scheduled backup jobs in the cluster

• pve_list_cluster_backup_info

  List guests that are not covered by any backup job

• pve_list_cluster_replication

  List all replication jobs in the cluster

• pve_list_cluster_resources

  List all cluster resources (VMs, containers, storage, nodes) with optional type filter

• pve_list_domains

  List all authentication domains/realms (e.g. pam, pve, ldap, ad)

• pve_list_firewall_aliases

  List all cluster-level firewall aliases (named IP/CIDR entries)

• pve_list_firewall_ipsets

  List all cluster-level firewall IP sets

• pve_list_firewall_rules

  List all cluster-level firewall rules

• pve_list_groups

  List all user groups

• pve_list_ha_resources

  List all HA-managed resources in the cluster

• pve_list_ha_rules

  List HA rules (node-affinity and resource-affinity). HA rules replace the deprecated HA groups (PVE 9+)

• pve_list_lxc_containers

  List all LXC containers on a specific node

• pve_list_lxc_snapshots

  List all snapshots of an LXC container

• pve_list_networks

  List all network interfaces on a specific node

• pve_list_node_services

  List all system services and their status on a specific node

• pve_list_node_storage

  List available storage on a specific node with usage information

• pve_list_nodes

  List all nodes in the Proxmox VE cluster

• pve_list_pools

  List all resource pools in the cluster

• pve_list_qemu_snapshots

  List all snapshots of a QEMU VM

• pve_list_qemu_vms

  List all QEMU virtual machines on a specific node

• pve_list_roles

  List all available roles and their privileges

• pve_list_storage

  List all configured storage backends in the cluster

• pve_list_storage_content

  List the content (disk images, ISOs, templates, backups) of a specific storage on a node

• pve_list_user_tokens

  List the API tokens of a user

• pve_list_users

  List all users in the PVE access control system

• pve_manage_node_service

  Start, stop, restart, or reload a system service on a specific node

• pve_migrate_qemu_vm

  Migrate a QEMU VM to another node in the cluster

• pve_pull_oci_image

  Pull an OCI container image from a registry (e.g. Docker Hub, GHCR) into a storage's template cache for use as an LXC container template (PVE 9.1+). Returns a task ID

• pve_rollback_lxc_snapshot

  Rollback an LXC container to a previous snapshot state

• pve_rollback_qemu_snapshot

  Rollback a QEMU VM to a previous snapshot state

• pve_run_backupwrite action

  Run an immediate backup (vzdump) of one or more VMs/containers on a node

• pve_update_aclwrite action

  Update access control list — grant or revoke roles for users/groups on specific paths

• pve_update_cluster_optionswrite action

  Update cluster-wide datacenter options

• pve_update_firewall_optionswrite action

  Update the cluster-level firewall options (e.g. enable/disable firewall)

• pve_update_firewall_rulewrite action

  Update an existing cluster-level firewall rule by position

• pve_update_ha_resourcewrite action

  Update the HA configuration for an existing managed resource

• pve_update_ha_rulewrite action

  Update an existing HA rule. Requires Sys.Console. Use delete_settings to unset options

• pve_update_lxc_configwrite action

  Update the configuration of an LXC container

• pve_update_networkwrite action

  Update the configuration of a network interface on a node

• pve_update_poolwrite action

  Update a resource pool — add or remove VMs/containers and storage from the pool

• pve_update_qemu_configwrite action

  Update the configuration of a QEMU VM

• pve_update_userwrite action

  Update an existing user's properties

• pve_update_user_tokenwrite action

  Update an API token's comment, expiration, or privsep

• Code scan43 source files scanned25/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitystatic API keys via environment variables6/15
• Maintenancelast push 18 days ago15/15
• Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10

Proxmox VE MCP Server: common questions

Is Proxmox VE MCP Server safe?

Mostly: it is graded B (83/100). Read the Proxmox VE MCP Server safety report

How do I install Proxmox VE MCP Server?

It runs on your machine. Copy the Claude Code, Claude Desktop or Cursor config from the install section.

Does Proxmox VE MCP Server need an API key?

Yes. The registry entry asks for PVE_TOKEN_ID, PVE_TOKEN_SECRET.

Is Proxmox VE MCP Server maintained?

The last commit was 18 days ago (2026-09-02). The latest release is v0.8.1.