🤖 Agent Creator

Agent Creator is an AutoGen-powered multi-agent system built in Python.
It provides a framework where AI agents can be dynamically created, communicate via messages, and operate inside a simulated environment (world.py).

Dependencies are managed with uv for fast and reproducible setups.

---

📂 Project Structure

├── agent.py           # Defines individual agent logic and roles
├── creator.py         # Handles agent instantiation and orchestration
├── messages.py        # Defines message structures and communication utilities
├── world.py           # Main entry point – runs the AutoGen-powered world simulation
├── pyproject.toml     # Project dependencies & metadata
├── uv.lock            # Lockfile for uv dependency manager
├── .python-version    # Python version specification
├── .gitignore         # Files ignored by Git
└── README.md          # This documentation

---

🚀 Getting Started

1. Clone the repository

git clone https://github.com/ceodaniyal/Agent-Creator.git
cd Agent-Creator

2. Install dependencies

uv install

3. Run the project

uv run python world.py

This launches the AutoGen multi-agent simulation, where agents interact and exchange messages.

---

🔧 How It Works

• world.py → Main entry point. Initializes the AutoGen environment and runs the simulation.
• agent.py → Defines agent behavior and logic.
• creator.py → Dynamically creates and orchestrates agents.
• messages.py → Handles communication protocols between agents.

---

🛠️ Customization

• Add new agent types in agent.py
• Expand world rules and flows in world.py
• Extend or refine message handling in messages.py
• Use creator.py to test new orchestration strategies

---

🌟 Why Agent Creator?

• Built on AutoGen, making it easy to define, manage, and experiment with multi-agent systems.
• Modular structure for quick customization.
• Reproducible environments powered by uv.
• Perfect for research, prototyping, and simulations involving multiple AI agents.

---

🤝 Contributing

Contributions are welcome!

1. Fork the repo
2. Create a feature branch (feature/new-agent)
3. Submit a pull request

Do you also want me to add an example AutoGen workflow snippet (e.g., showing how two agents interact in world.py) inside the README so new users immediately see how it works?