Agentic Framework Using Dependency Injection?

[lovedeepatsgit]

I have tried Simple Plugins and ChatCompletionServices Using DI, and I liked this approach, Do anyone have any idea how We can Agentic Framework Using DI, Like there is , Is there anything for multiagent group chat, because it involves Selection and termination Strategies also? Any Recommendations, any previous samples or work done are most welcome.

using Microsoft.Extensions.Configuration;
using Microsoft.SemanticKernel;
using Microsoft.SemanticKernel.ChatCompletion;
using Microsoft.Extensions.DependencyInjection;
using DependencyInjectedKernel.DiPlugins;
using Microsoft.SemanticKernel.Connectors.AzureOpenAI;
using Microsoft.SemanticKernel.Connectors.OpenAI;
using Microsoft.Extensions.Logging;

public class Program
{
public static async Task Main(string[] args)
{
var configuration = new ConfigurationBuilder()
.SetBasePath(Directory.GetCurrentDirectory())
.AddJsonFile("appsettings.json", optional: false, reloadOnChange: true)
.AddJsonFile($"appsettings.{Environment.GetEnvironmentVariable("ASPNETCORE_ENVIRONMENT")}.json", optional: true)
.Build();

    string AZURE_OPEN_AI_ENDPOINT = configuration["LlmModels:AiFoundryEndpoint"];
    string AZURE_OPEN_AI_KEY = configuration["LlmModels:AiFoundryApiKey"];
    string DEPLOYMENT_MODEL = configuration["LlmModels:GPT-4o-2"];

    var builder = Kernel.CreateBuilder();
    builder.Services.AddLogging(b => b.AddConsole().SetMinimumLevel(LogLevel.Trace));
    Kernel kernel = builder.AddAzureOpenAIChatCompletion(deploymentName: DEPLOYMENT_MODEL,endpoint: AZURE_OPEN_AI_ENDPOINT,apiKey: AZURE_OPEN_AI_KEY).Build();

    kernel.ImportPluginFromType<DiPluginConnector>();

    var settings = new AzureOpenAIPromptExecutionSettings() { ToolCallBehavior = ToolCallBehavior.AutoInvokeKernelFunctions };
    var chatService = kernel.GetRequiredService<IChatCompletionService>();
    ChatHistory chat = new();

    while (true)
    {
        Console.Write("USER: ");
        chat.AddUserMessage(Console.ReadLine());

        var response = await chatService.GetChatMessageContentAsync(chat, settings, kernel);
        Console.WriteLine($"AI: {response}");
        chat.Add(response);
    }
}

}

[crickman]

For agents, you can use keyed-services as demonstrated in the sample: https://github.com/microsoft/semantic-kernel/blob/main/dotnet/samples/GettingStartedWithAgents/Step06_DependencyInjection.cs

For group-chat, I might consider defining a factory with one of the ServiceCollectionServiceExtensions