MATAC

MATAC System (Multi-Agent Task Allocation and Coordination)

MATAC is a Python-based system designed for multi-agent task allocation and coordination using natural language processing and state machine concepts.

Features

• Dynamic task allocation based on agent skills
• State-based agent management (idle, assigned, executing, completed)
• Asynchronous task execution
• Natural language task description and allocation
• Integration with OpenAI's GPT models for intelligent decision-making

Components

1. Agent: Represents individual agents with specific skills and state management.
2. TaskAllocator: Handles task assignment to the most suitable agent.
3. TaskSignature: Defines the structure for task descriptions and allocations.

Prerequisites

• Python 3.7+
• OpenAI API key

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/matac.git
   cd matac

2. Install dependencies:

   pip install dspy python-dotenv transitions

3. Set up environment variables: Create a `.env` file in the project root and add your OpenAI API key:

   OPENAI_API_KEY=your_api_key_here

Usage

1. Initialize agents with specific skills:

   agents = [
       Agent(llm, name="Agent 1", skills=["frontend", "backend"]),
       Agent(llm, name="Agent 2", skills=["backend"]),
       Agent(llm, name="Agent 3", skills=["frontend", "design"])
   ]

2. Create a TaskAllocator:

   task_allocator = TaskAllocator(llm, agents)

3. Define tasks and run the main execution loop:

   tasks_to_allocate = [
       "Build the user interface",
       "Set up the database",
       "Design the user experience"
   ]
   
   async def main():
       for task in tasks_to_allocate:
           task_allocator.allocate_task(task)
       await task_allocator.coordinate_execution()
   
   asyncio.run(main())

Customization

• Modify the Agent class to add more states or t