# Microsoft Agent Framework 1.0 — Production Multi-Agent Orchestration

> Microsoft unifies Semantic Kernel and AutoGen into one production agent framework with MCP and A2A support.

- Org: Microsoft
- Date: 2026-04-03
- Category: tool, repo
- Source: https://devblogs.microsoft.com/agent-framework/microsoft-agent-framework-version-1-0/
- Canonical page: https://ai-tldr.dev/releases/microsoft-agent-framework-1/

## In short

Production-ready agent framework unifying Semantic Kernel and AutoGen with MCP tool discovery, graph-based workflows, multi-provider LLM support, and a browser-based DevUI debugger. 9.4k stars.

## What it is

Microsoft Agent Framework 1.0 is a production-ready SDK for building AI agents and multi-agent workflows in Python and C#/.NET. It merges the capabilities of two popular Microsoft projects, Semantic Kernel and AutoGen, into a single framework with stable APIs and long-term support.

## How it works

Agents connect to LLMs from multiple providers (Azure OpenAI, Anthropic, Google, Ollama) through a unified interface. A graph-based workflow engine composes agents into deterministic pipelines with streaming, checkpointing, and human-in-the-loop gates. MCP support lets agents discover and invoke external tools dynamically, and A2A protocol support enables cross-framework agent collaboration.

## Why it matters

Enterprise teams building multi-agent systems no longer have to choose between Semantic Kernel's production polish and AutoGen's orchestration patterns. The framework's MCP and A2A support means agents built with it can interoperate with tools and agents from other ecosystems.

## Who it's for

Enterprise developers building production AI agent systems in .NET or Python.

## Try it

pip install agent-framework

## Links

- [Announcement](https://devblogs.microsoft.com/agent-framework/microsoft-agent-framework-version-1-0/)
- [Repo](https://github.com/microsoft/agent-framework)
- [Docs](https://learn.microsoft.com/en-us/agent-framework/)