Libvirt MCP

What is Libvirt MCP about?

Libvirt MCP is an experimental server that implements the Model Context Protocol (MCP) to bridge libvirt, the virtualization management library, with large language models hosted on Ollama. By doing so, it enables AI‑assisted operations such as creating, configuring, and monitoring virtual machines through natural‑language prompts.

How to use Libvirt MCP?

1. Clone and install the CLI tools

   git clone https://github.com/chrishayuk/mcp-cli
   pip3.11 install -e "[.][cli,dev]"
   

2. Install Ollama and pull a model

   curl -fsSL https://ollama.com/install.sh | sh
   ollama serve >/dev/null 2>&1 &
   ollama pull granite3.2:8b-instruct-q8_0
   

3. Install Python dependencies

   pip install uv
   dnf install -y libvirt-devel python3-devel
   uv sync
   

4. Configure the server
   • Edit server_config.json to point to the correct path of the libvirt‑mcp server.
5. Run the server

   ./run.sh
   

   run.sh launches the MCP server using Ollama as the provider and the Granite model.
6. Optional debugging

   dnf install -y npm
   pip install mcp
   mcp dev setup.py
   

Key features of Libvirt MCP

• MCP integration: Implements the Model Context Protocol for seamless AI interaction.
• Ollama backend: Uses locally hosted Ollama models, removing reliance on external APIs.
• Granite model support: Pre‑configured to work with the Granite 3.2 instruction‑tuned model.
• Simple setup: Scripted installation with uv sync and a single run.sh launch command.
• Debugging utilities: Ability to install the generic mcp package and run development mode.

Use cases of Libvirt MCP

• AI‑driven VM provisioning: Create and configure VMs by describing requirements in natural language.
• Automated troubleshooting: Ask the model to diagnose and fix common libvirt issues.
• Learning tool: Explore libvirt commands and concepts through conversational queries.
• Prototype AI operations: Quickly test AI‑enabled workflows for cloud infrastructure management.

FAQ from the Libvirt MCP

Q: Do I need an internet connection? A: Only for the initial Ollama installation and model download. After the model is cached locally, the server works offline.

Q: Which operating systems are supported? A: The instructions target Fedora/RHEL based systems (use dnf). The setup can be adapted to other Linux distributions by installing equivalent packages.

Q: Can I use a different LLM provider? A: The current run.sh is hard‑coded for Ollama, but the MCP protocol is provider‑agnostic; swapping the provider would require custom scripting.

Q: How do I update the model? A: Use ollama pull <model-name> to fetch a newer version, then restart run.sh.

Q: Is this production‑ready? A: It is experimental and intended for testing and prototyping rather than production deployments.

libvirt-mcp

This is an experimental mcp server for libvirt. The following lines explain how to use it with mcp-cli and ollama. First, install mcp-cli:

git clone https://github.com/chrishayuk/mcp-cli
pip3.11 install -e ".[cli,dev]"

Then, install ollama:

curl -fsSL https://ollama.com/install.sh | sh
ollama serve >/dev/null 2>&1  &
ollama pull granite3.2:8b-instruct-q8_0

You need also uv:

pip install uv

You need the following python bindings:

dnf install -y libvirt-devel python3-devel

Then, in the libvirt-mcp directory, first install the dependencies by running:

uv sync

Then, edit server_config.json and set up the correct path to the libvirt-mcp server. Finally, execute run.sh, that uses ollama as provider and granite as model.

For debugging, you can install mcp:

dnf install -y npm
pip install mcp

And then, run:

mcp dev setup.py