Microsoft has released a .NET SDK for the AG-UI protocol, designed to standardize communication between AI agents and front-end applications. The SDK is integrated into the Microsoft Agent Framework and supports bidirectional interaction between agents and .NET services.
The release of the Microsoft AG-UI .NET SDK provides a formal bridge between complex AI agent logic and the user-facing interfaces that streaming platforms use for discovery and support. By standardizing how agents emit tool calls and lifecycle events, Microsoft reduces the custom engineering required to deploy interactive AI assistants within the .NET ecosystem. This move aligns with a broader industry shift toward interoperable AI protocols, potentially lowering the barrier for streaming services to integrate sophisticated agents into their existing infrastructure. Watch for how quickly third-party developers adopt these five NuGet packages to build cross-platform agentic workflows.
Microsoft's Agent Framework achieved full AG-UI compatibility in November 2025, when CopilotKit announced that the Microsoft Agent Framework could connect directly to AG-UI frontends without custom socket code or polling loops. The integration was developed collaboratively between CopilotKit and Microsoft engineers, enabling developers to stream agent messages, tool calls, and state updates into responsive React components through a shared event schema. AG-UI now supports frameworks including LangGraph, CrewAI, Mastra, Google ADK, and Microsoft Agent Framework across both Python and .NET runtimes.
The Microsoft Agent Framework itself launched in public preview on October 1, 2025, as a unified open-source SDK converging AutoGen and Semantic Kernel into a single commercial-grade orchestration layer. The framework integrates with Azure AI Foundry for observability and governance, supports the Agent2Agent (A2A) protocol for cross-runtime collaboration, and connects to tools dynamically via Model Context Protocol (MCP). This positions the AG-UI .NET SDK release as a natural extension of Microsoft's broader agent infrastructure strategy, giving C# developers a first-party path to expose agent endpoints to front-end applications.
For developers evaluating agent-to-UI protocols, AG-UI occupies a distinct layer compared to MCP. While MCP handles agent-to-tool communication, AG-UI specifically addresses the streaming of responses, tool invocations, shared state updates, and UI component synchronization between agents and user interfaces. CopilotKit published a technical guide demonstrating how to build a front-end for Microsoft Agent Framework agents using AG-UI as the bridge layer, showing the adapter pattern where the .NET backend exposes an AG-UI endpoint via the Microsoft.Agents.AI.Hosting.AGUI.AspNetCore package and a CopilotKit runtime connects through an HttpAgent client class.
Microsoft released the AG-UI .NET SDK to standardize how AI agents communicate with front-end applications. Developed with CopilotKit, this tool enables C# developers to implement bidirectional, event-based interactions. By formalizing agent-to-UI protocols, it reduces custom engineering requirements, making it easier to integrate sophisticated AI assistants into existing .NET-based streaming and software infrastructures.
The SDK provides a standardized protocol for AI agents to interact with front-end applications, enabling bidirectional communication between agents and .NET services.
While MCP focuses on agent-to-tool communication, AG-UI is specifically designed for streaming responses, tool invocations, shared state updates, and synchronizing UI components between agents and user interfaces.
AG-UI supports several frameworks, including LangGraph, CrewAI, Mastra, Google ADK, and the Microsoft Agent Framework, across both Python and .NET runtimes.
The SDK is published on NuGet under the MIT license and includes five specific packages, such as AGUI.Protobuf for binary event stream formatting and AGUI.Client for HTTP integration.
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source