Add Microsoft Semantic Kernel implementations of agent architectures in Python and C#

[Copilot]

Duplicates the 8 LangChain/LangGraph agent architecture notebooks using Microsoft Semantic Kernel, providing Python and C# polyglot notebook implementations for framework comparison.

Changes

New folder: agent-architectures-maf/

• Python notebooks (.ipynb): 9 notebooks using semantic-kernel package
• C# polyglot notebooks (.dib): 9 notebooks using Microsoft.SemanticKernel NuGet
• Documentation: README with setup for both languages, requirements.txt, .env-template

Agent patterns implemented

All 8 architectures from original LangChain version:

1. Simple LLM call
2. Tool calling (manual + ReAct)
3. Routing
4. Reflection
5. Parallelism
6. Supervisor
7. Swarm
8. Planner

Framework mapping examples

Python - LangChain to Semantic Kernel:

# Before (LangChain)
from langchain_openai import AzureChatOpenAI
from langchain.tools import tool

llm = AzureChatOpenAI(model="gpt-4")
llm_with_tools = llm.bind_tools([weather_tool])

# After (Semantic Kernel)
from semantic_kernel.connectors.ai.open_ai import AzureChatCompletion
from semantic_kernel.functions import kernel_function

kernel = sk.Kernel()
kernel.add_service(AzureChatCompletion(...))
kernel.add_plugin(WeatherPlugin(), plugin_name="weather")
# Auto function calling via FunctionChoiceBehavior.Auto()

C# - Polyglot notebooks:

#r "nuget: Microsoft.SemanticKernel, 1.32.0"

public class WeatherPlugin {
    [KernelFunction, Description("Get weather")]
    public string GetWeather(string location) => ...
}

var kernel = Kernel.CreateBuilder()
    .AddAzureOpenAIChatCompletion(...)
    .Plugins.AddFromType<WeatherPlugin>()
    .Build();

Repository structure update

Updated main README.md to reference new folder alongside original LangChain implementations.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Duplicate Langchain Notebooks and make use of Microsoft Agent Framework</issue_title>
<issue_description>We need a copy of the exciting Langchain/Langgraph notebooks in the Architecture folder.

But instead of Langchain library it needs to make use of Microsoft Agent Framework in Python but also a version With polyglot notebooks in c#</issue_description>

Comments on the Issue (you are @copilot in this section)

@sammydeprez Assign to @copilot

• Fixes Duplicate Langchain Notebooks and make use of Microsoft Agent Framework #3

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.