ProxmoxMCP-Plus
MCP server for Proxmox VE: manage VMs, LXC, snapshots and backups from an agent with permission control and command approval
- 649 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Manages live virtualization: VMs, LXC, snapshots, backups
- Requires a Proxmox API token with mutating permissions
- Native MCP over HTTP can be exposed without auth if misconfigured
Install
- Works in
Manual install
uvx proxmox-mcp-plusRun MCP stdio from PyPI without a system install. Requires uv.
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Manages live virtualization: VMs, LXC, snapshots, backups
- Requires a Proxmox API token with mutating permissions
- Native MCP over HTTP can be exposed without auth if misconfigured
This is third-party code. Review the repository files before installing.
What it does
The server connects Proxmox VE to agents and HTTP tooling through one control layer. It covers day-two operations: virtual machines and LXC, snapshots, backups, ISOs, storage and cluster node queries. Long-running tasks are tracked by stable job_ids tied to the Proxmox UPID, with status polling, retry, cancel and history. Two paths are available: native MCP for agents and an OpenAPI bridge for dashboards and no-code flows. It installs via uvx or pip from PyPI, or runs in docker.
Who it is for. For devops engineers and homelab owners who run virtualization on Proxmox VE.
Good fit when
- You want to manage VMs and LXC on Proxmox from an agent
- You need snapshots, backups and storage operations on demand
- You need to track long-running Proxmox tasks with retry and cancel
Not a fit when
- You have no Proxmox API token with the right permissions
- You do not want to give an agent access to live virtualization
- You need read-only viewing without the risk of mutating operations
Example request
Show Proxmox nodes and VMs and take a snapshot of vm 101 before the updateMCP
- Transport
- stdio, http, sse
- Authentication
- key is optional
Environment variables
MCP_API_KEYsecret- Key for native MCP over HTTP, sent in the Authorization header and independent of Proxmox credentials
PROXMOX_MCP_MODE- Server run mode, for example mcp-http for native MCP over HTTP
Limitations
You need access to Proxmox VE and an API token; the token's permissions define what the server can do. Configuration lives in proxmox-config/config.json with the node address and token. Native MCP over HTTP needs a separate MCP_API_KEY and an Authorization header on every request, otherwise access must be opened explicitly with an unsafe option. The server performs mutating operations against infrastructure.
How to disable. Remove the server entry from your MCP client config, stop the docker container if running, and revoke the Proxmox API token.
FAQ
How does native MCP differ from OpenAPI?
Native MCP is for agents. The OpenAPI bridge exposes the same operations over HTTP for dashboards, scripts and no-code tools.
How is Proxmox access configured?
Through proxmox-config/config.json, where you set the node address, port and an API token with the needed permissions.
How is ProxmoxMCP-Plus different from calling the Proxmox API directly?
The server gives the agent one Proxmox VE management layer with permission control and command approval. It covers VMs and LXC, snapshots, backups, ISOs and storage. Native MCP and HTTP tools through OpenAPI are available.
How do I set up Proxmox access for MCP?
Create a separate API token in Proxmox with minimal rights and set it in the config. Run uvx proxmox-mcp-plus. It works with Claude Desktop, Cursor, VS Code and Codex. Risk is high because the server manages VMs.
Discussion
No questions or reviews yet.
Related
MCP serversA Proxmox VE MCP server with 341 tools: virtual machines, containers, storage, cluster and backups, with a read-only mode.
An MCP server for Proxmox VE, Backup Server, Mail Gateway and Datacenter Manager: plan before action, a change ledger and undo
A PowerShell module and skill for Claude Code and Codex: the agent finds, starts and snapshots Proxmox VE VMs via the API.
A self-hosted homelab hub with a built-in MCP server: the agent gets 32 tools limited by per-key API scopes
A self-hosted gateway that merges several MCP servers into one endpoint with auth and tool selection
The official Prometheus MCP server: PromQL queries, metrics, rules and alerts plus built-in runbooks for troubleshooting.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| ProxmoxMCP-Plus (this card) | no | key is optional | 649 | |
| kubectl-mcp-server | no | not required | 972 | reads cluster state and diagnoses pod failures |
| Unreal Engine MCP Server | no | not required | 1,032 | Unreal Engine assets, actors and blueprints |
| Android MCP | no | not required | 924 | controls an Android device through ADB |
README in short
The README covers driving Proxmox VE from MCP clients, agents and OpenAPI tools with a security focus: Proxmox API tokens, OpenAPI bearer auth, command policy, approval tokens and TLS validation. It shows three run paths: MCP stdio via uvx proxmox-mcp-plus, native MCP over HTTP via docker compose, and the OpenAPI bridge. Long-running operations are tracked by job_id and Proxmox UPID. A code execution mode with three tools in an isolated sandbox is documented separately. MIT licensed.