feat: externalize AI configuration, implement resilience policies, and update extraction prompt formatting

backlog-ai.md

@@ -1,6 +1,6 @@
 # 🤖 LLM Agent Implementation Backlog: AI Semantic Integration
 
-**Project Context:** .NET 10, EF Core (SQLite), `Microsoft.Extensions.AI`.
+**Project Context:** .NET 10, EF Core (SQLite), `Microsoft.Extensions.AI`, [`GeminiDotnet`](https://github.com/rabuckley/GeminiDotnet).
 **Core Goal:** Integrate Gemini 1.5 Flash with a persistent Semantic Cache to minimize API costs and latency.
 
 ---
@@ -31,22 +31,22 @@
 ---
 
 ## 🧠 Phase 2: AI Client & Contract Definition
-**Objective:** Set up the communication bridge with Google Gemini API.
+**Objective:** Set up the communication bridge with Google Gemini API using [`GeminiDotnet`](https://github.com/rabuckley/GeminiDotnet).
 
 ### Task 2.1: Define Data Transfer Objects (DTOs)
-* **Target Folder:** `Core/DTOs/AI`.
+* **Target Folder:** `NexusReader.Application/DTOs/AI`.
 * **Requirements:**
     * Define `KnowledgePacket` record containing `List<KeyConcept>` and `List<QuizQuestion>`.
     * Use `[JsonPropertyName]` attributes for strict JSON mapping.
 * **LLM Instructions:** "Define immutable records for the AI response schema. Ensure they match the expected JSON structure from the system prompt."
 
 ### Task 2.2: Infrastructure AI Client Setup
-* **Target:** `Program.cs` / Dependency Injection.
+* **Target:** `NexusReader.Infrastructure/DependencyInjection.cs`.
 * **Requirements:**
-    * Install `Microsoft.Extensions.AI` and `Microsoft.Extensions.AI.Google`.
+    * Install `Microsoft.Extensions.AI` and `GeminiDotnet.Extensions.AI`.
-    * Register `IChatClient` using `GoogleChatClient`.
+    * Register `IChatClient` using `GeminiChatClient`.
     * Inject `ApiKey` from `IConfiguration`.
-* **LLM Instructions:** "Register the GoogleChatClient in the DI container. Use the .NET 10 `AddChatClient` extension pattern."
+* **LLM Instructions:** "Register the `GeminiChatClient` in the DI container. Use the .NET 10 `AddChatClient` extension pattern."
 
 ---
 

src/NexusReader.Infrastructure/Configuration/AiSettings.cs

@@ -0,0 +1,13 @@
+namespace NexusReader.Infrastructure.Configuration;
+
+public class AiSettings
+{
+    public const string SectionName = "Ai:Google";
+
+    public string ApiKey { get; set; } = string.Empty;
+    public string Model { get; set; } = "gemini-1.5-flash";
+    public int MaxInputLength { get; set; } = 15000;
+    public int MaxOutputTokens { get; set; } = 1000;
+    public int RetryAttempts { get; set; } = 3;
+    public double Temperature { get; set; } = 0.1;
+}

src/NexusReader.Infrastructure/DependencyInjection.cs

@@ -7,6 +7,9 @@ using GeminiDotnet.Extensions.AI;
 using NexusReader.Infrastructure.Persistence;
 using NexusReader.Application.Abstractions.Services;
 using NexusReader.Infrastructure.Services;
+using NexusReader.Infrastructure.Configuration;
+using Polly;
+using Polly.Retry;
 
 namespace NexusReader.Infrastructure;
 
@@ -18,17 +21,31 @@ public static class DependencyInjection
         services.AddDbContext<AppDbContext>(options =>
             options.UseSqlite(connectionString));
 
-        var apiKey = configuration["Ai:Google:ApiKey"];
+        services.Configure<AiSettings>(configuration.GetSection(AiSettings.SectionName));
-        if (string.IsNullOrWhiteSpace(apiKey))
+        var aiSettings = configuration.GetSection(AiSettings.SectionName).Get<AiSettings>() ?? new AiSettings();
-        {
-            throw new InvalidOperationException("AI Studio ApiKey is missing in configuration (Ai:Google:ApiKey).");
-        }
-        var modelId = configuration["Ai:Google:Model"] ?? "gemini-1.5-flash";
 
-        services.AddSingleton<IChatClient>(new GeminiChatClient(new GeminiClientOptions 
+        if (string.IsNullOrWhiteSpace(aiSettings.ApiKey) || aiSettings.ApiKey == "PLACEHOLDER")
+        {
+            // We don't throw here to allow the app to start, but services using AI will fail gracefully
+        }
+
+        services.AddResiliencePipeline("ai-retry", builder =>
+        {
+            builder.AddRetry(new RetryStrategyOptions
+            {
+                ShouldHandle = new PredicateBuilder().Handle<Exception>(ex => 
+                    ex.Message.Contains("429") || ex.Message.Contains("Too Many Requests") || ex.Message.Contains("quota")),
+                BackoffType = DelayBackoffType.Exponential,
+                UseJitter = true,
+                MaxRetryAttempts = aiSettings.RetryAttempts,
+                Delay = TimeSpan.FromSeconds(2)
+            });
+        });
+
+        services.AddChatClient(new GeminiChatClient(new GeminiClientOptions 
         { 
-            ApiKey = apiKey, 
+            ApiKey = aiSettings.ApiKey, 
-            ModelId = modelId 
+            ModelId = aiSettings.Model 
         }));
 
         services.AddScoped<IKnowledgeService, KnowledgeService>();

src/NexusReader.Infrastructure/NexusReader.Infrastructure.csproj

@@ -12,6 +12,7 @@
     </PackageReference>
     <PackageReference Include="Microsoft.EntityFrameworkCore.Sqlite" Version="10.0.7" />
     <PackageReference Include="Microsoft.Extensions.AI" Version="10.5.0" />
+    <PackageReference Include="Microsoft.Extensions.Resilience" Version="10.5.0" />
     <PackageReference Include="Polly" Version="8.6.6" />
     <PackageReference Include="Polly.Extensions.Http" Version="3.0.0" />
     <PackageReference Include="VersOne.Epub" Version="3.3.6" />

src/NexusReader.Infrastructure/Services/KnowledgeService.cs

@@ -8,7 +8,9 @@ using NexusReader.Domain.Entities;
 using NexusReader.Infrastructure.Helpers;
 using NexusReader.Infrastructure.Persistence;
 using Polly;
-using Polly.Retry;
+using Polly.Registry;
+using Microsoft.Extensions.Options;
+using NexusReader.Infrastructure.Configuration;
 
 namespace NexusReader.Infrastructure.Services;
 
@@ -16,25 +18,20 @@ public class KnowledgeService : IKnowledgeService
 {
     private readonly IChatClient _chatClient;
     private readonly AppDbContext _dbContext;
+    private readonly ResiliencePipeline _retryPipeline;
+    private readonly AiSettings _settings;
     private const string PromptVersion = "1.0";
-    private const string ModelId = "gemini-1.5-flash";
 
-    private static readonly ResiliencePipeline _retryPipeline = new ResiliencePipelineBuilder()
+    public KnowledgeService(
-        .AddRetry(new RetryStrategyOptions
+        IChatClient chatClient, 
-        {
+        AppDbContext dbContext,
-            ShouldHandle = new PredicateBuilder().Handle<Exception>(ex => 
+        ResiliencePipelineProvider<string> pipelineProvider,
-                ex.Message.Contains("429") || ex.Message.Contains("Too Many Requests") || ex.Message.Contains("quota")),
+        IOptions<AiSettings> settings)
-            BackoffType = DelayBackoffType.Exponential,
-            UseJitter = true,
-            MaxRetryAttempts = 3,
-            Delay = TimeSpan.FromSeconds(2)
-        })
-        .Build();
-
-    public KnowledgeService(IChatClient chatClient, AppDbContext dbContext)
     {
         _chatClient = chatClient;
         _dbContext = dbContext;
+        _retryPipeline = pipelineProvider.GetPipeline("ai-retry");
+        _settings = settings.Value;
     }
 
     public async Task<Result<KnowledgePacket>> GetKnowledgeAsync(string text, CancellationToken cancellationToken = default)
@@ -48,10 +45,9 @@ public class KnowledgeService : IKnowledgeService
         var normalizedText = ContentHasher.Normalize(text);
 
         // Phase 4: Request Pre-processing (Token Saving)
-        const int MaxInputLength = 15000; // Roughly 3k-4k tokens
+        if (normalizedText.Length > _settings.MaxInputLength)
-        if (normalizedText.Length > MaxInputLength)
         {
-            return Result.Fail($"Input text is too long ({normalizedText.Length} characters after normalization). Max allowed is {MaxInputLength}.");
+            return Result.Fail($"Input text is too long ({normalizedText.Length} characters after normalization). Max allowed is {_settings.MaxInputLength}.");
         }
 
         // Simple token estimation (4 chars per token)
@@ -86,8 +82,8 @@ public class KnowledgeService : IKnowledgeService
             var options = new ChatOptions
             {
                 ResponseFormat = ChatResponseFormat.Json,
-                Temperature = 0.1f,
+                Temperature = (float)_settings.Temperature,
-                MaxOutputTokens = 1000
+                MaxOutputTokens = _settings.MaxOutputTokens
             };
 
             var response = await _retryPipeline.ExecuteAsync(async ct => 
@@ -117,7 +113,7 @@ public class KnowledgeService : IKnowledgeService
             {
                 ContentHash = hash,
                 JsonData = jsonResponse,
-                ModelId = ModelId,
+                ModelId = _settings.Model,
                 PromptVersion = PromptVersion,
                 CreatedAt = DateTime.UtcNow
             };

src/NexusReader.Infrastructure/Services/PromptRegistry.cs

@@ -4,6 +4,6 @@ public static class PromptRegistry
 {
     public const string KnowledgeExtractionSystemPrompt = 
         "You are an expert educator. Analyze the provided text to extract key concepts and generate relevant quizzes. " +
-        "Return ONLY a minified JSON object that strictly adheres to the provided schema. No markdown formatting, no explanations. " +
+        "CRITICAL: Return ONLY a minified JSON object. Do NOT include markdown formatting like ```json or ```. Do NOT include explanations. " +
         "Schema: { \"concepts\": [ { \"title\": \"string\", \"description\": \"string\" } ], \"quizzes\": [ { \"question\": \"string\", \"options\": [ \"string\" ], \"correct_index\": 0 } ] }.";
 }