DevExpress Reports - Generate a Report Based on a User Prompt Within a Console App

This example integrates multi-agent report generation for a .NET 8 console application using Azure OpenAI.

The generated report is saved to the generatedReport.repx file.

Prerequisites

• .NET 8 SDK
• Azure OpenAI
• The following NuGet packages should be installed in your project:
  • DevExpress.AIIntegration.Reporting.Common
  • Azure.AI.OpenAI
  • Microsoft.Extensions.AI

Implementation Details

To implement report generation in your application, you must:

1. Create an IChatClient instance for your AI provider:

   // Retrieve the Azure OpenAI endpoint, key, and model from user environment variables.
   string endpoint = Environment.GetEnvironmentVariable("AZURE_OPENAI_ENDPOINT", EnvironmentVariableTarget.User)
       ?? throw new InvalidOperationException("AZURE_OPENAI_ENDPOINT is not set.");
   string apiKey = Environment.GetEnvironmentVariable("AZURE_OPENAI_API_KEY", EnvironmentVariableTarget.User)
       ?? throw new InvalidOperationException("AZURE_OPENAI_API_KEY is not set.");
   string deploymentName = Environment.GetEnvironmentVariable("AZURE_OPENAI_DEPLOYMENT", EnvironmentVariableTarget.User)
       ?? "gpt-5-mini";
   // Create an Azure OpenAI client to work with the chat agent.
   IChatClient chatClient = new AzureOpenAIClient(new Uri(endpoint), new ApiKeyCredential(apiKey))
       .GetChatClient(deploymentName)
       .AsIChatClient();

   For additional information, refer to the following help topic: AI-powered Extensions — Register AI Clients.

2. Initialize an AI extensions container to register the AI client and reporting extension:

   using DevExpress.AIIntegration.Reporting.Common.Extensions;
   using DevExpress.AIIntegration;
   //
   AIExtensionsContainerDefault container = AIExtensionsContainerConsole.CreateDefaultAIExtensionContainer(chatClient);
   container.RegisterReportingExtensions();

3. Create a class that implements IAIReportGenerationHost to handle clarification questions and to show progress/notifications:

   Implement and supply a host for interactive workflows (ConsoleAIReportGenerationHost.cs).

   namespace Reporting.Generation.Console {
       public class ConsoleAIReportGenerationHost : IAIReportGenerationHost {
           private string lastStatus = string.Empty;
   
           public Task<PromptClarificationAnswer> ClarifyPromptAsync(PromptClarificationQuestion request) {
               // Render the request in your UI (request.Text and request.Choices).
               // Return PromptClarificationAnswer.FromValue(selectedChoice).
   
               // Return PromptClarificationAnswer.Canceled() if the user cancels the operation.
               return Task.FromResult(PromptClarificationAnswer.Canceled());
           }
   
           public void NotifyAsync(string status, string reasoning) {
               bool isNewStatus = lastStatus != status;
               lastStatus = status;
   
               // Surface progress to users (console, logger, status bar, web socket, etc.).
               // Update the status when isNewStatus is true; otherwise refresh the reasoning only.
           }
       }
   }

4. Generate the report from a prompt.

   In the Program.cs file, create a PromptToReportRequest instance with the user prompt, assign the host, and specify additional settings. Call AIReportingIntegration.GeneratePromptToReportAsync to obtain an XtraReport instance.

   using Reporting.Generation.Console;
   using DevExpress.AIIntegration.Reporting.Common.Extensions;
   using DevExpress.XtraReports.UI;
   // ...
   // Ask the user for a report description in natural language.
   Console.WriteLine("Specify a prompt to generate the report:");
   string prompt = Console.ReadLine();
   try {
       // Create a host to handle clarification questions and progress notifications.
       ConsoleAIReportGenerationHost host = new ConsoleAIReportGenerationHost();
       // Build a generation request from the user prompt.
       PromptToReportRequest generationRequest = new PromptToReportRequest(userPrompt: prompt, dataSourceSchema: null, report: null) {
           ReportGenerationHost = host,
           FixLayoutErrors = true
       };
       // Generate a report layout and save it to a REPX file.
       XtraReport report = await container.GeneratePromptToReportAsync(generationRequest, default);
       report.SaveLayoutToXml("generatedReport.repx");
   } catch (Exception ex) {
       Console.WriteLine($"Report generation failed: {ex.Message}");
   }

Files to Review

• ConsoleAIReportGenerationHost.cs/(VB: ConsoleAIReportGenerationHost.vb)
• Program.cs/(VB: Program.vb)

Documentation

• PromptToReportRequest Class
• IAIReportGenerationHost Interface
• PromptClarificationQuestion Class
• PromptClarificationAnswer Class

Does This Example Address Your Development Requirements/Objectives?

(you will be redirected to DevExpress.com to submit your response)