VME MCP CLI

Rich terminal chat interface for VME infrastructure management with voice support.

Installation

# Basic installation
pip install vme-mcp-cli

# With audio support for voice commands
pip install "vme-mcp-cli[audio]"

MCP Server Requirement

The VME MCP CLI requires an MCP server to communicate with your VME/Morpheus infrastructure. The server package (vme-mcp-server) is not yet available on PyPI.

For installation instructions, see the main repository.

Remote Server Option

If you have a remote MCP server running, configure it in your config file:

server:
  servers:
    vme:
      transport: http
      path_or_url: http://your-server:8080

Quick Start

# 1. Create default configuration
vme-mcp-cli config

# 2. Set your API credentials
export VME_API_BASE_URL=https://your-vme.com/api
export VME_API_TOKEN=your-token
export ANTHROPIC_API_KEY=your-anthropic-key
export OPENAI_API_KEY=your-openai-key  # Optional: for voice features

# 3. Start the chat interface
vme-mcp-cli

# Or with debug output
vme-mcp-cli --debug-level 3

Configuration

Config file location:

• Linux/Mac: ~/.config/vme-cli/config.yaml
• Windows: %APPDATA%\vme-cli\config.yaml

Default Configuration

llm:
  anthropic_key: ${ANTHROPIC_API_KEY}
  default_provider: anthropic
  default_model: claude-3-5-sonnet-20241022

server:
  servers:
    vme:
      name: "vme"
      transport: "stdio"
      command: "vme-mcp-server"  # Requires server package installed
      env:
        VME_API_BASE_URL: ${VME_API_BASE_URL}
        VME_API_TOKEN: ${VME_API_TOKEN}
        VME_LAZY_LOADING: "true"  # Fast startup with on-demand tool loading

audio:
  enabled: false
  voice: "alloy"  # OpenAI voice options: alloy, echo, fable, onyx, nova, shimmer

ui:
  theme: github_dark
  show_thinking_indicator: true

Environment Variables

The CLI supports environment variable substitution in the config file:

# Required
VME_API_BASE_URL=https://your-vme.com/api
VME_API_TOKEN=your-api-token
ANTHROPIC_API_KEY=your-anthropic-key

# Optional
OPENAI_API_KEY=your-openai-key  # For voice features
VME_LAZY_LOADING=true           # Control server startup behavior
VME_PRELOAD_GROUPS=compute      # Preload specific tool groups

Features

• 🎨 Rich Terminal Interface - GitHub-inspired dark theme with syntax highlighting
• 💬 Natural Language Processing - Chat with your infrastructure using plain English
• 🎙️ Voice Commands - Speak to manage VMs (requires OpenAI API key)
• 🔧 Progressive Tool Discovery - Start with 18 tools, unlock 500+ on demand
• 📊 Real-time Status - Visual indicators for thinking, speaking, and operations
• 🔒 Secure Credentials - Environment variables and encrypted config support
• ⚡ Lazy Loading - 70% faster startup with on-demand tool creation
• 🤖 Multi-LLM Support - Works with Anthropic Claude and OpenAI GPT

Usage Examples

Basic Commands

> What tools are available?
> Show me the appliance settings
> Get current license information

VM Management

> List all VMs
> Create a VM called web-server-01
> Show me VMs in zone tc-lab
> Delete VM with ID 123

Progressive Discovery

> Discover compute infrastructure    # Unlocks 40+ VM tools
> Discover networking capabilities   # Unlocks network tools
> Discover storage capabilities      # Unlocks storage tools

Advanced Operations

> Create a Ubuntu 22.04 VM with 4 CPUs and 8GB RAM
> Show me all available instance types
> What virtual images are available?
> Help me create a web server with specific requirements

Requirements

• Python 3.10+
• Anthropic API key (for Claude LLM)
• VME/Morpheus API credentials (URL and token)
• OpenAI API key (optional, for voice features)
• PortAudio (optional, for microphone support)

Audio Setup (Optional)

For voice command support:

pip install "vme-mcp-cli[audio]"

Linux users: Install PortAudio first:

• Ubuntu/Debian: sudo apt-get install libportaudio2
• Fedora/RHEL: sudo dnf install portaudio

Windows/macOS: No additional steps needed (PortAudio included automatically)

Troubleshooting

"MCP server not found" Error

The server package is required but not yet available on PyPI. See the installation guide for setup instructions.

Audio Not Working

1. Ensure PortAudio is installed (see Audio Setup above)
2. Set your OpenAI API key: export OPENAI_API_KEY=your-key
3. Enable audio in config: audio.enabled: true
4. Check microphone permissions

Documentation

• Full Documentation - Installation, server setup, and development
• VME Infrastructure Guide - Understanding VME/Morpheus
• Report Issues - Bug reports and feature requests

License

See LICENSE in the main repository.