Multi-Agent Patterns in Python

This repository contains practical examples demonstrating multi-agent orchestration patterns using the Microsoft Agent Framework.

Overview

Multi-agent systems enable complex task decomposition by coordinating multiple specialized AI agents. This repository showcases five distinct orchestration patterns, each with working Python examples that you can run and adapt for your own use cases.

Patterns

Pattern	Description
01-SequentialAgents	Chain agents where output flows from one to the next
02-ConcurrentAgents	Run multiple agents in parallel and aggregate results
03-HandoffAgents	Triage agents that route to specialized handlers
04-GroupChatAgents	Multi-agent discussions with manager or selector-based coordination
05-MagenticAgents	Advanced orchestration with planning, checkpointing, and human-in-the-loop

Prerequisites

• Python 3.10 or later
• Azure OpenAI resource with a deployed model (e.g., gpt-4o or gpt-4.1)
• Azure CLI installed and authenticated (az login)

Quick Start

1. Clone the repository

   git clone https://github.com/YOUR_USERNAME/maf-multi-agent-patterns-python.git
   cd maf-multi-agent-patterns-python

2. Create and activate a virtual environment

   python -m venv venv
   
   # Windows
   venv\Scripts\activate
   
   # Linux/Mac
   source venv/bin/activate

3. Install dependencies

   pip install -r requirements.txt

4. Configure environment variables

   cp .env.example .env

   Edit .env with your Azure OpenAI settings:

   AZURE_OPENAI_ENDPOINT="https://your-resource.openai.azure.com/"
   AZURE_OPENAI_CHAT_DEPLOYMENT_NAME="gpt-4o"

5. Authenticate with Azure

   az login

6. Run a sample

   cd 01-SequentialAgents
   python sequential_agents.py

Project Structure

maf-multi-agent-patterns-python/
├── .env.example              # Environment variable template
├── .gitignore                # Git ignore rules
├── LICENSE                   # MIT License
├── README.md                 # This file
├── requirements.txt          # Python dependencies
├── 01-SequentialAgents/      # Sequential pattern examples
├── 02-ConcurrentAgents/      # Concurrent pattern examples
├── 03-HandoffAgents/         # Handoff pattern examples
├── 04-GroupChatAgents/       # Group chat pattern examples
└── 05-MagenticAgents/        # Magentic pattern examples

Environment Variables

Variable	Description	Required
AZURE_OPENAI_ENDPOINT	Azure OpenAI resource endpoint	Yes
AZURE_OPENAI_CHAT_DEPLOYMENT_NAME	Model deployment name (e.g., gpt-4o)	Yes
AZURE_OPENAI_API_VERSION	API version (default: 2024-08-01-preview)	No
AZURE_AI_PROJECT_ENDPOINT	Azure AI Foundry project endpoint (for code interpreter samples)	Some samples

Attribution

The sample code in this repository is based on and adapted from the Microsoft Agent Framework examples. The Agent Framework is an open-source SDK for building AI agents with Azure OpenAI and other model providers.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Resources

• Azure AI Agents Documentation
• Multi-Agent Design Patterns
• Azure AI Foundry