Semantic Kernel–Agent Framework

In this post I show you the recently introduced Semantic Kernel agents feature and how it simplifies building your own AI agents. But maybe I should start with a short recap about Semantic Kernel.

On the documentation pages, Semantic Kernel is described like this:

Semantic Kernel is a lightweight, open-source development kit that lets you easily build AI agents and integrate the latest AI models into your C#, Python, or Java codebase. It serves as an efficient middleware that enables rapid delivery of enterprise-grade solutions.

It gives you all the building blocks required to build your own agent; a chat completion model, a plugin system, a planner and more. However until recently you had to bring all this building blocks together yourself.

Here is a small code snippet I copied from an existing project:

	// Import packages
	using Microsoft.SemanticKernel;
	using Microsoft.SemanticKernel.ChatCompletion;
	using Microsoft.SemanticKernel.Connectors.OpenAI;
	// Create a kernel with Azure OpenAI chat completion
	var builder = Kernel.CreateBuilder().AddAzureOpenAIChatCompletion(modelId, endpoint, apiKey);
	// Add enterprise components
	builder.Services.AddLogging(services => services.AddConsole().SetMinimumLevel(LogLevel.Trace));
	// Build the kernel
	Kernel kernel = builder.Build();
	var chatCompletionService = kernel.GetRequiredService<IChatCompletionService>();
	// Initialize plugin and add to the kernel
	kernel.Plugins.AddFromType<MenuPlugin>("Menu");
	// Enable auto invocation of plugins
	OpenAIPromptExecutionSettings openAIPromptExecutionSettings = new()
	{
	ToolCallBehavior = ToolCallBehavior.AutoInvokeKernelFunctions
	};
	// Create a history store the conversation
	var history = new ChatHistory("You are a waiter answering questions about the menu.");
	// Initiate a back-and-forth chat
	string? userInput;
	do {
	// Collect user input
	Console.Write("User > ");
	userInput = Console.ReadLine();
	// Add user input
	history.AddUserMessage(userInput);
	// Get the response from the AI
	var result = await chatCompletionService.GetChatMessageContentAsync(
	history,
	executionSettings: openAIPromptExecutionSettings,
	kernel: kernel);
	// Print the results
	Console.WriteLine("Assistant > " + result);
	// Add the message from the agent to the chat history
	history.AddAssistantMessage(result.Content ?? string.Empty);
	} while (userInput is not null);

view raw SemanticKernel.cs hosted with ❤ by GitHub

There are a lot of things going on in the code above and if you have hard time to understand all of this I have some good news for you. Starting with the Python (1.6.0) and .NET releases (1.18.0 RC1), Semantic Kernel now provides a first-class abstraction for agents.

To use it, we first need to add the following NuGet package:

dotnet add package Microsoft.SemanticKernel.Agents.Core

Let’s rewrite the code above to use the new agent abstraction:

	// Import packages
	using Microsoft.SemanticKernel;
	using Microsoft.SemanticKernel.ChatCompletion;
	using Microsoft.SemanticKernel.Connectors.OpenAI;
	// Create a kernel with Azure OpenAI chat completion
	var builder = Kernel.CreateBuilder().AddAzureOpenAIChatCompletion(modelId, endpoint, apiKey);
	// Add enterprise components
	builder.Services.AddLogging(services => services.AddConsole().SetMinimumLevel(LogLevel.Trace));
	// Build the kernel
	Kernel kernel = builder.Build();
	// Define the agent
	ChatCompletionAgent agent =
	new()
	{
	Instructions = "Answer questions about the menu.",
	Name = "Waiter",
	Kernel = kernel,
	Arguments = new KernelArguments(new OpenAIPromptExecutionSettings() { ToolCallBehavior = ToolCallBehavior.AutoInvokeKernelFunctions }),
	};
	// Initialize plugin and add to the agent's Kernel (same as direct Kernel usage).
	KernelPlugin plugin = KernelPluginFactory.CreateFromType<MenuPlugin>();
	agent.Kernel.Plugins.Add(plugin);
	// Create the chat history to capture the agent interaction.
	ChatHistory chat = [];
	// Initiate a back-and-forth chat
	string? userInput;
	do
	{
	// Collect user input
	Console.Write("User > ");
	userInput = Console.ReadLine();
	// Add user input
	chat.AddUserMessage(userInput);
	// Get the response from the AI
	#pragma warning disable SKEXP0110 // Type is for evaluation purposes only and is subject to change or removal in future updates. Suppress this diagnostic to proceed.
	await foreach (ChatMessageContent response in agent.InvokeAsync(chat))
	{
	// Print the results
	Console.WriteLine("Assistant > " + response.ToString());
	// Add the message from the agent to the chat history
	chat.Add(response);
	}
	#pragma warning restore SKEXP0110 // Type is for evaluation purposes only and is subject to change or removal in future updates. Suppress this diagnostic to proceed.
	} while (userInput is not null);

view raw Agent.cs hosted with ❤ by GitHub

This is already an improvement but you still have to manage the chat history yourself.

If you are using an OpenAI based model, you can go one step further and use the OpenAI assistant abstraction so that the state is managed for you:

	//Define the agent
	OpenAIAssistantAgent agent =
	await OpenAIAssistantAgent.CreateAsync(
	kernel: new(),
	config: new("<ApiKey>", "<Endpoint URL>"),
	new()
	{
	Name = "Waiter",
	Instructions = "Answer questions about the menu.",
	ModelId = "<ModelId>",
	EnableCodeInterpreter = false,
	});
	// Initialize plugin and add to the agent's Kernel (same as direct Kernel usage).
	KernelPlugin plugin = KernelPluginFactory.CreateFromType<MenuPlugin>();
	agent.Kernel.Plugins.Add(plugin);
	// Initiate a back-and-forth chat
	string threadId = await agent.CreateThreadAsync();
	string? userInput;
	do
	{
	// Collect user input
	Console.Write("User > ");
	userInput = Console.ReadLine();
	await agent.AddChatMessageAsync(threadId, new ChatMessageContent(AuthorRole.User, userInput));
	// Get the response from the AI
	await foreach (ChatMessageContent message in agent.InvokeAsync(threadId))
	{
	// Print the results
	Console.WriteLine("Assistant > " + message.ToString());
	}
	} while (userInput is not null);

view raw OpenAIAssistant.cs hosted with ❤ by GitHub

Nice!

Remark: Everything I showing here is still in preview and will probably change in the future.

More information

semantic-kernel/dotnet/samples/GettingStartedWithAgents at main · microsoft/semantic-kernel (github.com)

Introducing enterprise multi-agent support in Semantic Kernel | Semantic Kernel (microsoft.com)