how do I register Microsoft.SemanticKernel.Text Generation.ITextGenerationService?

[anna-trivino]

Hello. I am just starting out with Semantic Kernel and trying things out.

I've implemented Native plugins in my sample project. And they are running fine. Now, I am trying to incorporate prompt plugins into my code but during execution of the prompt plugin, I get this error:
Microsoft.SemanticKernel.TextGeneration.ITextGenerationService not registered.

Hoping for some guidance about this.

I have my code set up this way:

DI

`

    //filters
    services.AddTransient<PromptModerationFilter>();
    services.AddTransient<ExceptionHandlingFilter>();
    services.AddTransient<AutoFunctionInvocationFilter>();

    //ai integrations
    services.AddAzureOpenAIChatCompletion(
        deploymentName: configuration["AzureOpenAiGpt4:DeploymentName"]!,
        endpoint: configuration["AzureOpenAiGpt4:Endpoint"]!,
        apiKey: configuration["AzureOpenAiGpt4:ApiKey"]!,
        serviceId: "AzureOpenAiGpt4"
    );


    services.AddTransient<Kernel>(sp => {

        Kernel kernel = new(sp);
        
        //add filters
        kernel.FunctionInvocationFilters.Add(sp.GetRequiredService<ExceptionHandlingFilter>());
        kernel.AutoFunctionInvocationFilters.Add(sp.GetRequiredService<AutoFunctionInvocationFilter>());

        //add native plugins
        kernel.Plugins.AddFromObject(sp.GetRequiredService<UpdateChatNamePlugin>());

        //add prompt plugins
        kernel.ImportPluginFromPromptDirectoryYaml(Path.Combine(AppDomain.CurrentDomain.BaseDirectory, "Plugins", "PokemonPlugin"), "PokemonPlugin", promptTemplateFactory: new HandlebarsPromptTemplateFactory());

        return kernel;

    });

`

Chat

`

        public async IAsyncEnumerable<string> GetStreamingChatResponseAsync(ChatHistory chatHistory, ModelType modelType, [EnumeratorCancellation]  CancellationToken cancellationToken)
          {
          var chatCompletion = _kernel.GetRequiredService<IChatCompletionService>(modelType.ToString());
  
          var settings = new AzureOpenAIPromptExecutionSettings
        {
            ToolCallBehavior = ToolCallBehavior.AutoInvokeKernelFunctions
        };

          await foreach (var responseChunk in chatCompletion.GetStreamingChatMessageContentsAsync(chatHistory, settings, _kernel, cancellationToken))
              {
            string chunkText = responseChunk.Content!; 

            yield return chunkText; 

        }
    }

`

Exception Handling Filter (from: https://devblogs.microsoft.com/semantic-kernel/filters-in-semantic-kernel/)

`
using Microsoft.Extensions.Logging;
using Microsoft.SemanticKernel;

  namespace Test;
  
  public class ExceptionHandlingFilter : IFunctionInvocationFilter
  {
      private readonly ILogger<ExceptionHandlingFilter> _logger;
      public ExceptionHandlingFilter(ILogger<ExceptionHandlingFilter> logger)
      {
          _logger = logger;
      }

  public async Task OnFunctionInvocationAsync(FunctionInvocationContext context, Func<FunctionInvocationContext, Task> next)
  {
      try
      {
          await next(context);
      }
      catch (Exception exception)
      {
          this._logger.LogError(exception, "Something went wrong during function invocation");

          // Example: override function result value
          context.Result = new FunctionResult(context.Result, "Friendly message instead of exception");

          // Example: Rethrow another type of exception if needed
          // throw new InvalidOperationException("New exception");
      }
  }
}

`

[markwallace-microsoft]

@rogerbarreto can you take a look at this one, it looks like:

1. IChatCompletionService is not being resolved
2. Error message should mentioned chat completion or text generation

[rogerbarreto]

@anna-trivino Please take a look in the answer here.

• .Net: how do I register Microsoft.SemanticKernel.Text Generation.ITextGenerationService? #11197

TLDR;
Ensure when using serviceId identifiers with AddAzureOpenAIChatCompletion to also provide the ServiceId in the PromptExecutionSettings.
i.e:

var settings = new PromptExecutionSettings { ServiceId = "AzureOpenAiGpt4" };
var result = kernel.InvokePromptAsync("my prompt", new(settings));

[a-trivino]

Thank you!