Can I use this with Azure OpenAI?

Yes, it provides full support for Azure OpenAI with various authentication methods, including Azure CLI and Bearer Tokens.

What is the Microsoft Agent Framework?

It is a framework for building, orchestrating, and deploying AI agents and multi-agent workflows in .NET, offering features like graph-based logic and streaming.

Can I persist conversation states between sessions?

Absolutely; the framework supports saving and restoring conversation states to JSON for use in stateless services or database persistence.

Does it support observability for production use?

Yes, it integrates with OpenTelemetry to provide detailed tracing and monitoring of agent activities, performance, and middleware pipelines.

How does this skill handle human-in-the-loop interactions?

It includes built-in patterns for pausing workflows to seek human approval before executing specific tool calls or sensitive functions.

Microsoft Agent Framework for .NET

Name: Microsoft Agent Framework for .NET
Author: Salmanferozkhan

bySalmanferozkhan

Data Science & ML

Builds and orchestrates sophisticated AI agents and multi-agent workflows using the Microsoft Agent Framework for .NET applications.

About

The Microsoft Agent Framework skill provides expert guidance for developers building autonomous AI agents and complex multi-agent systems within the .NET ecosystem. It enables advanced capabilities such as graph-based workflows, streaming, checkpointing, and human-in-the-loop interactions, making it ideal for creating reliable, stateful AI applications. Whether you are integrating OpenAI, Azure OpenAI, or Google Gemini, this skill helps you implement function tools, structured outputs, and observable middleware pipelines while following industry best practices for .NET development and Model Context Protocol (MCP) integration.

Key Features

Graph-based multi-agent orchestration and workflow design
Seamless integration with OpenAI, Azure OpenAI, and Google Gemini
Built-in observability using OpenTelemetry and custom middleware
Human-in-the-loop tool approvals and interactive workflows
State persistence with conversation checkpointing and time-travel capability
0 GitHub stars

Use Cases

Building enterprise chatbots with memory and stateful multi-turn conversations
Developing complex backend workflows that require human approval for sensitive actions
Creating autonomous research agents using function calling and tool execution

About

Key Features

Graph-based multi-agent orchestration and workflow design
Seamless integration with OpenAI, Azure OpenAI, and Google Gemini
Built-in observability using OpenTelemetry and custom middleware
Human-in-the-loop tool approvals and interactive workflows
State persistence with conversation checkpointing and time-travel capability
0 GitHub stars

Use Cases

Building enterprise chatbots with memory and stateful multi-turn conversations
Developing complex backend workflows that require human approval for sensitive actions
Creating autonomous research agents using function calling and tool execution