Skip to content

Microsoft Launches Open-Source Agent Framework for AI Agent Development

Unify and extend AI agent development with Microsoft's new Agent Framework. Streamline deployment across diverse environments and interact with a wide range of platforms.

There is a poster in which there is a robot, there are animated persons who are operating the...
There is a poster in which there is a robot, there are animated persons who are operating the robot, there are artificial birds flying in the air, there are planets, there is ground, there are stars in the sky, there is watermark, there are numbers and texts.

Microsoft Launches Open-Source Agent Framework for AI Agent Development

Microsoft has launched the Microsoft Agent Framework, an open-source SDK and runtime environment for developing, orchestrating, and deploying AI agents and complex multi-agent workflows. This new tool supports both .NET and Python, unifying and extending concepts from the Semantic Kernel and AutoGen projects.

The Agent Framework offers numerous features to streamline agent development and deployment. It supports various orchestration patterns and includes a package for experimental features, such as pluggable memory modules compatible with diverse storage solutions.

Agents can run in containers, on-premises, or across multiple clouds, ensuring portability across different environments. The framework enables the creation of both simple chat agents and complex multi-agent workflows with graph-based orchestration.

Any REST API with an OpenAPI specification can be instantly imported as a callable tool using the Agent Framework. Agents can interact with a wide range of platforms via Azure Logic Apps' integrated connections, including Azure services, Office 365 services, database servers, and enterprise systems.

Developers can use the framework locally and deploy applications to the Azure AI Foundry, which includes features for observability, fault tolerance, and compliance. Agents can dynamically discover and invoke external tools or data servers via the Model Context Protocol (MCP), and collaborate across different runtime environments using structured messaging controlled via the Agent2Agent (A2A) protocol. The framework is OpenAPI-first designed and cloud-independent.

The Microsoft Agent Framework simplifies AI agent development and deployment, offering extensive features and compatibility. It supports numerous patterns, experimental features, and seamless interaction with various platforms. With its support for .NET and Python, and its unifying approach to Semantic Kernel and AutoGen concepts, the Agent Framework is a powerful tool for developers.

Read also:

Latest