medula-cli 1.0.2

Command-line interface for Medula AI Agent Platform

Medula CLI

A command-line interface for the Medula AI Agent Platform, designed for developers who want to manage their AI agents from the terminal.

Installation

From PyPI (when published)

pip install medula-cli

Development Installation

git clone https://github.com/Subomi-olagoke/studious-rotary-phone
cd studious-rotary-phone
pip install -e .

Quick Install (Local)

pip install -r requirements.txt
chmod +x medula_cli.py
ln -s $(pwd)/medula_cli.py /usr/local/bin/medula

Quick Start

1. Login

medula auth login
# Enter your Medula email and password

2. List Your Agents

medula agents list

3. Create a New Agent

medula agents create --name "Support Bot" --model claude-3-haiku

4. Chat with Your Agent

medula chat start <agent-id>

5. Initialize a New Project

medula init my-ai-project
cd my-ai-project

Commands

Authentication

medula auth login               # Login to Medula platform
medula auth logout              # Logout and clear credentials  
medula auth whoami              # Show current user info

Agent Management

medula agents list              # List all your agents
medula agents create            # Create a new agent
medula agents show <agent-id>   # Show agent details
medula agents list --format json  # Output as JSON

Interactive Chat

medula chat start <agent-id>           # Start interactive chat
medula chat start <agent-id> --save-session name  # Save conversation

Chat commands while in session:

• /exit - End the conversation
• /save - Save current conversation
• /help - Show help

Training Data

medula data upload <agent-id> --text "Training content"
medula data upload <agent-id> --file document.pdf  # Coming soon

Project Management

medula init <project-name>      # Initialize new project

Configuration

The CLI stores configuration in ~/.medula/:

• config.json - General configuration
• Secure token storage via system keyring

Configuration Options

{
  "endpoint": "https://your-medula-instance.com",
  "tenant_id": "your-tenant-id", 
  "user_email": "[e-mail]"
}

Examples

Developer Workflow

# Initialize new project
medula init customer-support-bot
cd customer-support-bot

# Create agent
medula agents create --name "Customer Support" --model claude-3-sonnet

# Upload training data
medula data upload <agent-id> --text "Return Policy: Items can be returned within 30 days..."

# Test the agent
medula chat start <agent-id>

CI/CD Integration

# In your CI pipeline
export MEDULA_TOKEN="your-api-token"
medula agents list --format json | jq '.[] | select(.status=="active")'

Advanced Usage

Multiple Environments

# Development
medula auth login --endpoint https://dev-api.medula.ai

# Production  
medula auth login --endpoint https://api.medula.ai

JSON Output for Scripting

# Get all agents as JSON
medula agents list --format json

# Parse with jq
medula agents list --format json | jq '.[] | select(.model=="claude-3-haiku")'

API Integration

The CLI leverages your existing Medula API:

• POST /login - Authentication
• GET /api/v1/tenants/{tenant_id}/agents - List agents
• POST /api/v1/tenants/{tenant_id}/agents - Create agents
• POST /api/v1/tenants/{tenant_id}/agents/{agent_id}/chat - Chat

Troubleshooting

Authentication Issues

# Clear stored credentials
medula auth logout
medula auth login

# Check current status
medula auth whoami

Connection Issues

# Verify endpoint
medula auth whoami

# Test with different endpoint
medula auth login --endpoint https://your-instance.com

Debug Mode

# Enable verbose output
export MEDULA_DEBUG=1
medula agents list

Development

Project Structure

cli/
├── medula_cli.py      # Main CLI application
├── requirements.txt   # Dependencies
├── setup.py          # Package setup
└── README.md         # This file

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

License

MIT License - see LICENSE file for details.