Unraid
Query, monitor, and manage Unraid servers via GraphQL API through MCP tools. Supports system info, Docker, VMs, array/parity, notifications, plugins, rclone, and live telemetry.
What is Unraid?
Unraid is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to query, monitor, and manage unraid servers via graphql api through mcp tools. supports system info, docker, vms, array/parity, notifications, plugins, rclone, and live telemetry.
Query, monitor, and manage Unraid servers via GraphQL API through MCP tools. Supports system info, Docker, VMs, array/parity, notifications, plugins, rclone, and live telemetry.
This server falls under the Cloud Services category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Query, monitor, and manage Unraid servers via GraphQL API th
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx unraidConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Unraid
The Unraid MCP Server connects AI assistants to Unraid NAS and homelab servers via their GraphQL API, giving agents the ability to inspect system health, manage Docker containers and VMs, control array and parity operations, handle notifications, manage plugins, configure Rclone cloud remotes, and stream live telemetry such as CPU usage and log tails. All operations route through a single unified 'unraid' MCP tool using action and subaction parameters across 14 action groups and over 100 subactions. Homelab operators and self-hosters use it to query and control their Unraid infrastructure conversationally without opening the web UI.
Prerequisites
- An Unraid server with the GraphQL API accessible (typically https://tower.local/graphql)
- An Unraid API key generated from Unraid web UI under Settings → Management Access → API Keys
- Python 3.11+ and uv installed, OR Docker for containerised deployment
- An MCP-compatible client such as Claude Desktop or Claude Code
Clone the repository and configure environment
Clone the unraid-mcp repo and create your .env file from the provided template.
git clone https://github.com/jmagar/unraid-mcp.git
cd unraid-mcp
cp .env.example .envSet required environment variables
Edit .env and provide your Unraid server's GraphQL endpoint and your API key.
# .env
UNRAID_API_URL=https://tower.local/graphql
UNRAID_API_KEY=your_api_key_hereRun the server
Start the MCP server with uv. The server defaults to streamable-http transport on port 6970.
uv sync --dev
uv run unraid-mcp-serverAdd to your MCP client configuration
Configure your MCP client to connect to the running server via HTTP, or use stdio transport for a direct process connection.
{
"mcpServers": {
"unraid": {
"type": "streamable-http",
"url": "http://localhost:6970/mcp"
}
}
}Verify connectivity
Ask your AI assistant to check the Unraid connection. This exercises the health action group and confirms the API key is valid.
Unraid Examples
Client configuration
Connect Claude Desktop to a locally running unraid-mcp server via streamable-http.
{
"mcpServers": {
"unraid": {
"type": "streamable-http",
"url": "http://localhost:6970/mcp"
}
}
}Prompts to try
Commands that exercise Unraid's different action groups through the AI assistant.
- "Give me a system overview of my Unraid server."
- "List all running Docker containers and their status."
- "Start the Plex container."
- "What is the current CPU and memory usage?"
- "Show me the last 50 lines of the system log."
- "Is my parity check running, and what's the progress?"Troubleshooting Unraid
Connection refused or timeout when reaching the Unraid GraphQL API.
Confirm UNRAID_API_URL is correct and that the Unraid server is reachable from the machine running the MCP server. If using a hostname like 'tower.local', ensure mDNS resolves it or substitute the IP address.
API authentication errors even with a valid key.
Regenerate the API key in Unraid web UI under Settings → Management Access → API Keys. Paste the new key exactly into UNRAID_API_KEY in your .env file with no extra whitespace.
Destructive operations (array stop, VM force-stop) fail with a confirmation error.
All destructive actions require confirm=True to be explicitly passed. When using the AI assistant, explicitly state 'yes, confirm' in your request so the model passes the confirmation parameter.
Frequently Asked Questions about Unraid
What is Unraid?
Unraid is a Model Context Protocol (MCP) server that query, monitor, and manage unraid servers via graphql api through mcp tools. supports system info, docker, vms, array/parity, notifications, plugins, rclone, and live telemetry. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Unraid?
Follow the installation instructions on the Unraid GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Unraid?
Unraid works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Unraid free to use?
Yes, Unraid is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Unraid Alternatives — Similar Cloud Services Servers
Looking for alternatives to Unraid? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Unraid in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Unraid?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.