This repository provides a comprehensive set of hands-on examples and tutorials for building intelligent agents using the Microsoft Agent Framework. It targets engineers and developers looking to leverage AI agents, offering practical demonstrations for both Python and .NET implementations, covering agent foundations, tool integration, RAG, and multi-agent systems.

How It Works

The framework supports integration with multiple AI model providers, including Azure OpenAI, GitHub Models, and Microsoft Foundry. It facilitates the implementation of advanced agent capabilities such as vision analysis, code interpretation, custom tool usage, and Retrieval-Augmented Generation (RAG) for knowledge enhancement. The samples showcase agent orchestration, including multi-agent systems, sequential and concurrent workflows, and communication via the Model Context Protocol (MCP).

Quick Start & Requirements

• Installation:
  • Python: pip install agent-framework --pre (released) or pip install -r ./Installation/requirement.txt -U (build from source).
  • .NET: dotnet add package Microsoft.Agents.AI (released) or build from source via dotnet build agent-framework-dotnet.slnx.
• Prerequisites: Python 3.10+, .NET 9.0+, Visual Studio 2022 or VS Code with C# extension. Requires Azure OpenAI Service, Microsoft Foundry, GitHub Models, Azure CLI, and Azure Developer CLI (authenticated). Windows ARM64 may need specific OpenSSL configuration.
• Setup: Requires a .env file with API tokens and endpoints for various services.
• Links: Microsoft Agent Framework Documentation, Microsoft Foundry, GitHub Models.

Highlighted Details

• Multiple Provider Support: Azure OpenAI, GitHub Models, Microsoft Foundry.
• Tool Integration: Vision, code interpretation, custom functions.
• RAG Capabilities: File search and knowledge base integration.
• Multi-Agent Orchestration: Sequential and collaborative agent patterns.
• MCP Integration: Model Context Protocol for enhanced capabilities.
• Evaluation & Debugging: DevUI for visual debugging and observability tools for tracing.

Maintenance & Community

The repository welcomes contributions. Specific details on maintainers, sponsorships, or dedicated community channels (like Discord/Slack) are not explicitly detailed in the README.

Licensing & Compatibility

This project is licensed under the MIT License, which generally permits broad use, modification, and distribution, including for commercial purposes, with attribution.

Limitations & Caveats

Examples are primarily based on GitHub Models and Microsoft Foundry. Setup requires configuring multiple external service endpoints and API keys. Windows ARM64 users may encounter specific dependency configurations for OpenSSL. The framework has recently released v1.0, indicating ongoing development and potential for evolving sample implementations.