playbooks
Learn Cursor Rules Windsurf Rules MCPs Modes
Log in

Proxmox MCP server

Integrates with Proxmox virtualization environments to enable monitoring and management of virtual machines, containers, storage pools, and cluster resources through token-based authentication and structured API commands.
Back to servers
Setup instructions
Provider
Kevin
Release date
Feb 19, 2025
Language
Python
Stats
129 stars
View on GitHub

The Proxmox MCP Server provides a streamlined interface for managing Proxmox hypervisors through the Model Context Protocol (MCP), allowing you to interact with nodes, VMs, and containers through a clean Python-based API.

Installation

Prerequisites

Before you begin, make sure you have:

  • Python 3.10 or higher
  • Git
  • UV package manager (recommended)
  • Access to a Proxmox server with API token credentials

Also, gather the following information:

  • Proxmox server hostname or IP
  • Proxmox API token (token ID and value)

Setting Up the Server

  1. Clone the repository and set up the environment:
# Clone repository to your preferred directory
cd your/preferred/directory
git clone https://github.com/canvrno/ProxmoxMCP.git
cd ProxmoxMCP

# Create and activate virtual environment
uv venv
source .venv/bin/activate  # Linux/macOS
# OR
.\.venv\Scripts\Activate.ps1  # Windows
  1. Install the required dependencies:
# Install with development dependencies
uv pip install -e ".[dev]"
  1. Create and configure the server settings:
# Create config directory and copy template
mkdir -p proxmox-config
cp config/config.example.json proxmox-config/config.json
  1. Edit the configuration file proxmox-config/config.json:
{
    "proxmox": {
        "host": "PROXMOX_HOST",        // Your Proxmox server address
        "port": 8006,                  // Default is 8006
        "verify_ssl": false,           // Set false for self-signed certs
        "service": "PVE"               // Default is PVE
    },
    "auth": {
        "user": "USER@pve",            // Your Proxmox username
        "token_name": "TOKEN_NAME",    // API token ID
        "token_value": "TOKEN_VALUE"   // API token value
    },
    "logging": {
        "level": "INFO",               // Use DEBUG for more detail
        "format": "%(asctime)s - %(name)s - %(levelname)s - %(message)s",
        "file": "proxmox_mcp.log"      // Optional: Log to file
    }
}

Obtaining API Token

To set up a Proxmox API token:

  1. Log into your Proxmox web interface
  2. Navigate to Datacenter → Permissions → API Tokens
  3. Create a new API token:
    • Select a user (e.g., root@pam)
    • Enter a token ID (e.g., "mcp-token")
    • Uncheck "Privilege Separation" for full access
    • Save and copy both the token ID and secret value

Verifying Installation

Confirm that your installation is working properly:

# Check Python environment
python -c "import proxmox_mcp; print('Installation OK')"

# Run the server with your config file
# Linux/macOS
PROXMOX_MCP_CONFIG="proxmox-config/config.json" python -m proxmox_mcp.server

# Windows (PowerShell)
$env:PROXMOX_MCP_CONFIG="proxmox-config\config.json"; python -m proxmox_mcp.server

You should see either a successful connection to your Proxmox server or a connection error (if your Proxmox details are incorrect).

Running the Server

Development Mode

For testing and development:

# Activate virtual environment first
source .venv/bin/activate  # Linux/macOS
# OR
.\.venv\Scripts\Activate.ps1  # Windows

# Run the server
python -m proxmox_mcp.server

Cline Desktop Integration

If you're using Cline, add this configuration to your MCP settings file (typically at ~/.config/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json):

{
    "mcpServers": {
        "github.com/canvrno/ProxmoxMCP": {
            "command": "/absolute/path/to/ProxmoxMCP/.venv/bin/python",
            "args": ["-m", "proxmox_mcp.server"],
            "cwd": "/absolute/path/to/ProxmoxMCP",
            "env": {
                "PYTHONPATH": "/absolute/path/to/ProxmoxMCP/src",
                "PROXMOX_MCP_CONFIG": "/absolute/path/to/ProxmoxMCP/proxmox-config/config.json",
                "PROXMOX_HOST": "your-proxmox-host",
                "PROXMOX_USER": "username@pve",
                "PROXMOX_TOKEN_NAME": "token-name",
                "PROXMOX_TOKEN_VALUE": "token-value",
                "PROXMOX_PORT": "8006",
                "PROXMOX_VERIFY_SSL": "false",
                "PROXMOX_SERVICE": "PVE",
                "LOG_LEVEL": "DEBUG"
            },
            "disabled": false,
            "autoApprove": []
        }
    }
}

To generate the correct paths automatically, you can use:

# This will print the MCP settings with your absolute paths filled in
python -c "import os; print(f'''{{
    \"mcpServers\": {{
        \"github.com/canvrno/ProxmoxMCP\": {{
            \"command\": \"{os.path.abspath('.venv/bin/python')}\",
            \"args\": [\"-m\", \"proxmox_mcp.server\"],
            \"cwd\": \"{os.getcwd()}\",
            \"env\": {{
                \"PYTHONPATH\": \"{os.path.abspath('src')}\",
                \"PROXMOX_MCP_CONFIG\": \"{os.path.abspath('proxmox-config/config.json')}\",
                ...
            }}
        }}
    }}
}}''')"

Important notes:

  • All paths must be absolute
  • The Python interpreter must be from your virtual environment
  • The PYTHONPATH must point to the src directory
  • Restart VSCode after updating MCP settings

Available Tools

The server provides several MCP tools for interacting with Proxmox:

get_nodes

Lists all nodes in the Proxmox cluster.

  • Parameters: None
  • Example Response: 
    🖥️ Proxmox Nodes

🖥️ pve-compute-01
  • Status: ONLINE
  • Uptime: ⏳ 156d 12h
  • CPU Cores: 64
  • Memory: 186.5 GB / 512.0 GB (36.4%)

🖥️ pve-compute-02
  • Status: ONLINE
  • Uptime: ⏳ 156d 11h
  • CPU Cores: 64
  • Memory: 201.3 GB / 512.0 GB (39.3%)

get_node_status

Get detailed status of a specific node.

  • Parameters:
    • node (string, required): Name of the node
  • Example Response: 
    🖥️ Node: pve-compute-01
  • Status: ONLINE
  • Uptime: ⏳ 156d 12h
  • CPU Usage: 42.3%
  • CPU Cores: 64 (AMD EPYC 7763)
  • Memory: 186.5 GB / 512.0 GB (36.4%)
  • Network: ⬆️ 12.8 GB/s ⬇️ 9.2 GB/s
  • Temperature: 38°C

get_vms

List all VMs across the cluster.

  • Parameters: None
  • Example Response: 
    🗃️ Virtual Machines

🗃️ prod-db-master (ID: 100)
  • Status: RUNNING
  • Node: pve-compute-01
  • CPU Cores: 16
  • Memory: 92.3 GB / 128.0 GB (72.1%)

🗃️ prod-web-01 (ID: 102)
  • Status: RUNNING
  • Node: pve-compute-01
  • CPU Cores: 8
  • Memory: 12.8 GB / 32.0 GB (40.0%)

get_storage

List available storage.

  • Parameters: None
  • Example Response: 
    💾 Storage Pools

💾 ceph-prod
  • Status: ONLINE
  • Type: rbd
  • Usage: 12.8 TB / 20.0 TB (64.0%)
  • IOPS: ⬆️ 15.2k ⬇️ 12.8k

💾 local-zfs
  • Status: ONLINE
  • Type: zfspool
  • Usage: 3.2 TB / 8.0 TB (40.0%)
  • IOPS: ⬆️ 42.8k ⬇️ 35.6k

get_cluster_status

Get overall cluster status.

  • Parameters: None
  • Example Response: 
    ⚙️ Proxmox Cluster

  • Name: enterprise-cloud
  • Status: HEALTHY
  • Quorum: OK
  • Nodes: 4 ONLINE
  • Version: 8.1.3
  • HA Status: ACTIVE
  • Resources:
    - Total CPU Cores: 192
    - Total Memory: 1536 GB
    - Total Storage: 70 TB
  • Workload:
    - Running VMs: 7
    - Total VMs: 8
    - Average CPU Usage: 38.6%
    - Average Memory Usage: 42.8%

execute_vm_command

Execute a command in a VM's console using QEMU Guest Agent.

  • Parameters:
    • node (string, required): Name of the node where VM is running
    • vmid (string, required): ID of the VM
    • command (string, required): Command to execute
  • Example Response: 
    🔧 Console Command Result
  • Status: SUCCESS
  • Command: systemctl status nginx
  • Node: pve-compute-01
  • VM: prod-web-01 (ID: 102)

Output:
● nginx.service - A high performance web server and a reverse proxy server
   Loaded: loaded (/lib/systemd/system/nginx.service; enabled; vendor preset: enabled)
   Active: active (running) since Tue 2025-02-18 15:23:45 UTC; 2 months 3 days ago

Requirements:

  • VM must be running
  • QEMU Guest Agent must be installed and running in the VM
  • Command execution permissions must be enabled in the Guest Agent

How to install this MCP server

For Claude Code

To add this MCP server to Claude Code, run this command in your terminal:

claude mcp add-json "github-com-canvrno-ProxmoxMCP" '{"command":"/absolute/path/to/ProxmoxMCP/.venv/bin/python","args":["-m","proxmox_mcp.server"],"cwd":"/absolute/path/to/ProxmoxMCP","env":{"PYTHONPATH":"/absolute/path/to/ProxmoxMCP/src","PROXMOX_MCP_CONFIG":"/absolute/path/to/ProxmoxMCP/proxmox-config/config.json","PROXMOX_HOST":"your-proxmox-host","PROXMOX_USER":"username@pve","PROXMOX_TOKEN_NAME":"token-name","PROXMOX_TOKEN_VALUE":"token-value","PROXMOX_PORT":"8006","PROXMOX_VERIFY_SSL":"false","PROXMOX_SERVICE":"PVE","LOG_LEVEL":"DEBUG"},"disabled":false,"autoApprove":[]}'

See the official Claude Code MCP documentation for more details.

For Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > Tools & Integrations and click "New MCP Server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "github.com/canvrno/ProxmoxMCP": {
            "command": "/absolute/path/to/ProxmoxMCP/.venv/bin/python",
            "args": [
                "-m",
                "proxmox_mcp.server"
            ],
            "cwd": "/absolute/path/to/ProxmoxMCP",
            "env": {
                "PYTHONPATH": "/absolute/path/to/ProxmoxMCP/src",
                "PROXMOX_MCP_CONFIG": "/absolute/path/to/ProxmoxMCP/proxmox-config/config.json",
                "PROXMOX_HOST": "your-proxmox-host",
                "PROXMOX_USER": "username@pve",
                "PROXMOX_TOKEN_NAME": "token-name",
                "PROXMOX_TOKEN_VALUE": "token-value",
                "PROXMOX_PORT": "8006",
                "PROXMOX_VERIFY_SSL": "false",
                "PROXMOX_SERVICE": "PVE",
                "LOG_LEVEL": "DEBUG"
            },
            "disabled": false,
            "autoApprove": []
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explicitly ask the agent to use the tool by mentioning the tool name and describing what the function does.

For Claude Desktop

To add this MCP server to Claude Desktop:

1. Find your configuration file:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
  • Linux: ~/.config/Claude/claude_desktop_config.json

2. Add this to your configuration file:

{
    "mcpServers": {
        "github.com/canvrno/ProxmoxMCP": {
            "command": "/absolute/path/to/ProxmoxMCP/.venv/bin/python",
            "args": [
                "-m",
                "proxmox_mcp.server"
            ],
            "cwd": "/absolute/path/to/ProxmoxMCP",
            "env": {
                "PYTHONPATH": "/absolute/path/to/ProxmoxMCP/src",
                "PROXMOX_MCP_CONFIG": "/absolute/path/to/ProxmoxMCP/proxmox-config/config.json",
                "PROXMOX_HOST": "your-proxmox-host",
                "PROXMOX_USER": "username@pve",
                "PROXMOX_TOKEN_NAME": "token-name",
                "PROXMOX_TOKEN_VALUE": "token-value",
                "PROXMOX_PORT": "8006",
                "PROXMOX_VERIFY_SSL": "false",
                "PROXMOX_SERVICE": "PVE",
                "LOG_LEVEL": "DEBUG"
            },
            "disabled": false,
            "autoApprove": []
        }
    }
}

3. Restart Claude Desktop for the changes to take effect

Want to 10x your AI skills?

Get a free account and learn to code + market your apps using AI (with or without vibes!).

Nah, maybe later