Openstack MCP Server

What is Openstack MCP Server about?

The project implements a Model Context Protocol server that allows AI assistants to interact with an OpenStack cloud. It exposes tools for managing compute instances, flavors, images, identity/authentication, networks, and block storage through the MCP protocol.

How to use Openstack MCP Server?

Install Python 3.10+.
Ensure OpenStack credentials are defined in a clouds.yaml file.
Launch the server (examples provided for direct Python execution or via uvx).
Configure your AI assistant (e.g., Claude Desktop) to point to the MCP server, setting the OS_CLIENT_CONFIG_FILE environment variable to the path of clouds.yaml.

Key features of Openstack MCP Server

Full MCP protocol support for AI‑driven interactions.
Compute tools: create, list, delete servers and manage flavors.
Image tools: upload, list, delete OpenStack images.
Identity tools: handle authentication and project scopes.
Network tools: manage networks, subnets, routers, and ports.
Block storage tools: create, attach, detach, and delete volumes.

Use cases of Openstack MCP Server

Enabling conversational AI to provision or decommission virtual machines on demand.
Automating image lifecycle management through natural‑language commands.
Allowing AI assistants to configure networking and security groups without manual CLI steps.
Integrating AI‑driven operations into CI/CD pipelines for OpenStack‑based workloads.

FAQ from the Openstack MCP Server

What Python version is required?
- Python 3.10 or newer.
How do I provide OpenStack credentials?
- Create a clouds.yaml file with the required authentication fields and set OS_CLIENT_CONFIG_FILE to its location.
Can the server be run with a package manager?
- Yes, you can install the package via pip and then start it with python -m openstackmcp_server (or the provided entry point python-openstackmcp-server).
Is there support for other AI platforms besides Claude Desktop?
- Any MCP‑compatible client can connect; configuration simply needs the server’s command and environment variables.
Where can I contribute?
- See the CONTRIBUTING.rst file in the repository for guidelines.

python-openstackmcp-server

Openstack mcp server is a Model Context Protocol (MCP) server that provides an interface for AI assistants to interact with OpenStack services.

Table of Contents
Features
Quick Start with Claude Desktop

Features

MCP Protocol Support: Implements the Model Context Protocol for AI assistants.
Compute Tools: Manage OpenStack compute resources (servers, flavors).
Image Tools: Manage OpenStack images.
Identity Tools: Handle OpenStack identity and authentication.
Network Tools: Manage OpenStack networking resources.
Block Storage Tools: Manage OpenStack block storage resources.

Quick Start with Claude Desktop

Get started quickly with the OpenStack MCP server using Claude Desktop

Requirements

Python 3.10 or higher
OpenStack credentials configured in clouds.yaml file
Claude Desktop installed

Create or update your clouds.yaml file with your OpenStack credentials.

Example clouds.yaml:

clouds:
  openstack:
    auth:
      auth_url: https://your-openstack-auth-url.com
      username: your-username
      password: your-password
      project_name: your-project-name
      user_domain_name: Default
      project_domain_name: Default
    region_name: your-region
    interface: public
    identity_api_version: 3

Create or update your Claude Desktop configuration file:
- macOS: Edit $HOME/Library/Application Support/Claude/claude_desktop_config.json
- Windows: Edit %APPDATA%\Claude\claude_desktop_config.json
- Linux: Edit $HOME/.config/Claude/claude_desktop_config.json

Using python

{
  "mcpServers": {
    "openstack-mcp-server": {
      "command": "/path/to/your/python",
      "args": [
        "python-openstackmcp-server"
      ],
      "env" : {
        "OS_CLIENT_CONFIG_FILE": "/path/to/your/clouds.yaml"
      }
    }
  }
}

Using uvx

{
  "mcpServers": {
    "openstack-mcp-server": {
      "command": "uvx",
      "args": [
        "python-openstackmcp-server"
      ],
      "env" : {
        "OS_CLIENT_CONFIG_FILE": "/path/to/your/clouds.yaml"
      }
    }
  }
}

Development

Setup

This project supports both uv and tox for development and testing.

Using uv (Fast Local Development)

# Install dependencies (including dev and test groups)
uv sync

# Run tests
uv run --group test pytest

# Run linting
uv run ruff check src tests

# Format code
uv run ruff format src tests

Using tox (OpenStack Standard)

# Install tox
pip install tox
# or
uv tool install tox

# Run tests
tox -e py3

# Run linting
tox -e pep8

# Auto-format code
tox -e format

# Generate coverage report
tox -e cover

# Run arbitrary commands in virtualenv
tox -e venv -- <command>

# Test on specific Python version
tox -e py310  # or py311, py312, py313

# List all available environments
tox list

Testing

The project includes comprehensive test coverage (85%+). Tests are located in the tests/ directory.

# Run all tests
tox -e py3

# Run with coverage
tox -e cover

# Run with debugger
tox -e debug

# Run specific test file
tox -e py3 -- tests/tools/test_compute_tools.py

Contributing

Contributions are welcome! Please see the CONTRIBUTING file for details on how to contribute to this project.

License

This project is licensed under the Apache License 2.0. See the LICENSE file for details.

Openstack MCP Server