Microsoft Agent Framework Reaches Release Candidate, Unifying AI Agent Development

Microsoft has announced that the Microsoft Agent Framework has reached Release Candidate (RC) status for both .NET and Python, signaling a major step towards general availability.¹ This milestone indicates that the API surface is stable and feature-complete for what is planned in version 1.0, setting the stage for a production-ready toolset for developers building AI-powered assistants and complex agentic systems.

What is the Microsoft Agent Framework?

The Microsoft Agent Framework is an open-source development framework designed to build, orchestrate and deploy AI agents with a consistent programming model across .NET and Python.¹ It succeeds earlier efforts such as Semantic Kernel and AutoGen, consolidating agent creation, orchestration primitives, and multi-provider support under a single SDK.¹

Key Capabilities and Features

The framework offers a range of capabilities, including:

• Simple Agent Creation: Developers can create basic AI agents with just a few lines of code in either Python or .NET.¹
• Function Tools: Agents can call domain code in a type-safe manner.¹
• Graph-Based Workflows: The framework supports composing agents and functions into sequential, concurrent, and handoff patterns, with features like streaming, checkpointing, and human-in-the-loop support.¹
• Multi-Provider Support: It works with Microsoft Foundry, Azure OpenAI, OpenAI, GitHub Copilot, Anthropic Claude, AWS Bedrock, Ollama, and more.¹
• Interoperability: Supports standards such as MCP (Model Context Protocol) and agent-to-agent communication.¹
• Multi-Agent Orchestration: Developers can define workflows that coordinate multiple agents, enabling tasks with handoff logic and group chat patterns.¹

Code Examples

Here’s an example of creating a simple agent in .NET:

  using System; using Azure.Identity; using Microsoft.Agents.AI; var agent = new AzureOpenAIClient( new AzureCliCredential()) .GetChatClient("gpt-4.1") .AsAIAgent(name: "HaikuBot", instructions: "You are an upbeat assistant that writes beautifully."); Console.WriteLine(await agent.RunAsync("Write a haiku about Microsoft Agent Framework."));  

And here’s an example in Python:

  import asyncio from agent_framework.azure import AzureOpenAIResponsesClient from azure.identity import AzureCliCredential async def main(): agent = AzureOpenAIResponsesClient( credential=AzureCliCredential(), ).as_agent( name="HaikuBot", instructions="You are an upbeat assistant that writes beautifully." ) print(await agent.run("Write a haiku about Microsoft Agent Framework.")) asyncio.run(main())  

What’s Next?

As a Release Candidate, the packages are flagged as pre-release on NuGet and PyPI, and the framework is evolving based on early feedback.¹ Microsoft is preparing full GA documentation and migration guides, including guidance for teams migrating from Semantic Kernel or AutoGen. Developers should stay current with updates and be prepared for potential minor breaking changes until the General Availability release.¹

Resources for exploring the Agent Framework include the official documentation site, source code on GitHub, and migration aids.