Python: .Net: Dotnet devui compatibility fixes (#2026)

* DevUI: Add OpenAI Responses API proxy support with enhanced UI features

This commit adds support for proxying requests to OpenAI's Responses API,
allowing DevUI to route conversations to OpenAI models when configured to enable testing.

Backend changes:
- Add OpenAI proxy executor with conversation routing logic
- Enhance event mapper to support OpenAI Responses API format
- Extend server endpoints to handle OpenAI proxy mode
- Update models with OpenAI-specific response types
- Remove emojis from logging and CLI output for cleaner text

Frontend changes:
- Add settings modal with OpenAI proxy configuration UI
- Enhance agent and workflow views with improved state management
- Add new UI components (separator, switch) for settings
- Update debug panel with better event filtering
- Improve message renderers for OpenAI content types
- Update types and API client for OpenAI integration

* update ui, settings modal and workflow input form, add register cleanup hooks.

* add workflow HIL support, user mode, other fixes

* feat(devui): add human-in-the-loop (HIL) support with dynamic response schemas

Implement  HIL workflow support allowing workflows to pause for user input
with dynamically generated JSON schemas based on response handler type hints.

Key Features:
- Automatic response schema extraction from @response_handler decorators
- Dynamic form generation in UI based on Pydantic/dataclass response types
- Checkpoint-based conversation storage for HIL requests/responses
- Resume workflow execution after user provides HIL response

Backend Changes:
- Add extract_response_type_from_executor() to introspect response handlers
- Enrich RequestInfoEvent with response_schema via _enrich_request_info_event_with_response_schema()
- Map RequestInfoEvent to response.input.requested OpenAI event format
- Store HIL responses in conversation history and restore checkpoints

Frontend Changes:
- Add HILInputModal component with SchemaFormRenderer for dynamic forms
- Support Pydantic BaseModel and dataclass response types
- Render enum fields as dropdowns, strings as text/textarea, numbers, booleans, arrays, objects
- Display original request context alongside response form

Testing:
- Add  tests for checkpoint storage (test_checkpoints.py)
- Add schema generation tests for all input types (test_schema_generation.py)
- Validate end-to-end HIL flow with spam workflow sample

This enables workflows to seamlessly pause execution and request structured user input
with type-safe, validated forms generated automatically from response type annotations.

* improve HIL support, improve workflow execution view

* ui updates

* ui updates

* improve HIL for workflows, add auth and view modes

* update workflow

* security improvements , ui fixes

* fix mypy error

* update loading spinner in ui

* DevUI: Serialize workflow input as string to maintain conformance with OpenAI Responses format

* Phase 1: Add /meta endpoint and fix workflow event naming for .NET DevUI compatibility

* additional fixes for .NET DevUI workflow visualization item ID tracking

**Problem:**
.NET DevUI was generating different item IDs for ExecutorInvokedEvent and
ExecutorCompletedEvent, causing only the first executor to highlight in the
workflow graph. Long executor names and error messages also broke UI layout.

**Changes:**
- Add ExecutorActionItemResource to match Python DevUI implementation
- Track item IDs per executor using dictionary in AgentRunResponseUpdateExtensions
- Reuse same item ID across invoked/completed/failed events for proper pairing
- Add truncateText() utility to workflow-utils.ts
- Truncate executor names to 35 chars in execution timeline
- Truncate error messages to 150 chars in workflow graph nodes

** Details:**
- ExecutorActionItemResource registered with JSON source generation context
- Dictionary cleaned up after executor completion/failure to prevent memory leaks
- Frontend item tracking by unique item.id supports multiple executor runs
- All changes follow existing codebase patterns and conventions

Tested with review-workflow showing correct executor highlighting and state
transitions for sequential and concurrent executors.

* format fixes, remove cors tests

* remove unecessary attributes

---------

Co-authored-by: Mark Wallace <[email protected]>
Co-authored-by: Reuben Bond <[email protected]>

Author: 3 people

dotnet/src/Microsoft.Agents.AI.DevUI/DevUIExtensions.cs

@@ -32,6 +32,7 @@ public static IEndpointConventionBuilder MapDevUI(
     {
         var group = endpoints.MapGroup("");
         group.MapDevUI(pattern: "/devui");
+        group.MapMeta();
         group.MapEntities();
         return group;
     }

dotnet/src/Microsoft.Agents.AI.DevUI/Entities/EntitiesJsonContext.cs

@@ -15,10 +15,12 @@ namespace Microsoft.Agents.AI.DevUI.Entities;
     DefaultIgnoreCondition = JsonIgnoreCondition.WhenWritingNull)]
 [JsonSerializable(typeof(EntityInfo))]
 [JsonSerializable(typeof(DiscoveryResponse))]
+[JsonSerializable(typeof(MetaResponse))]
 [JsonSerializable(typeof(EnvVarRequirement))]
 [JsonSerializable(typeof(List<EntityInfo>))]
 [JsonSerializable(typeof(List<JsonElement>))]
 [JsonSerializable(typeof(Dictionary<string, JsonElement>))]
+[JsonSerializable(typeof(Dictionary<string, bool>))]
 [JsonSerializable(typeof(JsonElement))]
 [ExcludeFromCodeCoverage]
 internal sealed partial class EntitiesJsonContext : JsonSerializerContext;

dotnet/src/Microsoft.Agents.AI.DevUI/Entities/MetaResponse.cs

@@ -0,0 +1,66 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+using System.Text.Json.Serialization;
+
+namespace Microsoft.Agents.AI.DevUI.Entities;
+
+/// <summary>
+/// Server metadata response for the /meta endpoint.
+/// Provides information about the DevUI server configuration, capabilities, and requirements.
+/// </summary>
+/// <remarks>
+/// This response is used by the frontend to:
+/// - Determine the UI mode (developer vs user interface)
+/// - Check server capabilities (tracing, OpenAI proxy support)
+/// - Verify authentication requirements
+/// - Display framework and version information
+/// </remarks>
+internal sealed record MetaResponse
+{
+    /// <summary>
+    /// Gets the UI interface mode.
+    /// "developer" shows debug tools and advanced features, "user" shows a simplified interface.
+    /// </summary>
+    [JsonPropertyName("ui_mode")]
+    public string UiMode { get; init; } = "developer";
+
+    /// <summary>
+    /// Gets the DevUI version string.
+    /// </summary>
+    [JsonPropertyName("version")]
+    public string Version { get; init; } = "0.1.0";
+
+    /// <summary>
+    /// Gets the backend framework identifier.
+    /// Always "agent_framework" for Agent Framework implementations.
+    /// </summary>
+    [JsonPropertyName("framework")]
+    public string Framework { get; init; } = "agent_framework";
+
+    /// <summary>
+    /// Gets the backend runtime/language.
+    /// "dotnet" for .NET implementations, "python" for Python implementations.
+    /// Used by frontend for deployment guides and feature availability.
+    /// </summary>
+    [JsonPropertyName("runtime")]
+    public string Runtime { get; init; } = "dotnet";
+
+    /// <summary>
+    /// Gets the server capabilities dictionary.
+    /// Key-value pairs indicating which optional features are enabled.
+    /// </summary>
+    /// <remarks>
+    /// Standard capability keys:
+    /// - "tracing": Whether trace events are emitted for debugging
+    /// - "openai_proxy": Whether the server can proxy requests to OpenAI
+    /// </remarks>
+    [JsonPropertyName("capabilities")]
+    public Dictionary<string, bool> Capabilities { get; init; } = new();
+
+    /// <summary>
+    /// Gets a value indicating whether Bearer token authentication is required for API access.
+    /// When true, clients must include "Authorization: Bearer {token}" header in requests.
+    /// </summary>
+    [JsonPropertyName("auth_required")]
+    public bool AuthRequired { get; init; }
+}

dotnet/src/Microsoft.Agents.AI.DevUI/MetaApiExtensions.cs

@@ -0,0 +1,61 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+using Microsoft.Agents.AI.DevUI.Entities;
+
+namespace Microsoft.Agents.AI.DevUI;
+
+/// <summary>
+/// Provides extension methods for mapping the server metadata endpoint to an <see cref="IEndpointRouteBuilder"/>.
+/// </summary>
+internal static class MetaApiExtensions
+{
+    /// <summary>
+    /// Maps the HTTP API endpoint for retrieving server metadata.
+    /// </summary>
+    /// <param name="endpoints">The <see cref="IEndpointRouteBuilder"/> to add the route to.</param>
+    /// <returns>The <see cref="IEndpointConventionBuilder"/> for method chaining.</returns>
+    /// <remarks>
+    /// This extension method registers the following endpoint:
+    /// <list type="bullet">
+    /// <item><description>GET /meta - Retrieve server metadata including UI mode, version, capabilities, and auth requirements</description></item>
+    /// </list>
+    /// The endpoint is compatible with the Python DevUI frontend and provides essential
+    /// configuration information needed for proper frontend initialization.
+    /// </remarks>
+    public static IEndpointConventionBuilder MapMeta(this IEndpointRouteBuilder endpoints)
+    {
+        return endpoints.MapGet("/meta", GetMeta)
+            .WithName("GetMeta")
+            .WithSummary("Get server metadata and configuration")
+            .WithDescription("Returns server metadata including UI mode, version, framework identifier, capabilities, and authentication requirements. Used by the frontend for initialization and feature detection.")
+            .Produces<MetaResponse>(StatusCodes.Status200OK, contentType: "application/json");
+    }
+
+    private static IResult GetMeta()
+    {
+        // TODO: Consider making these configurable via IOptions<DevUIOptions>
+        // For now, using sensible defaults that match Python DevUI behavior
+
+        var meta = new MetaResponse
+        {
+            UiMode = "developer", // Could be made configurable to support "user" mode
+            Version = "0.1.0", // TODO: Extract from assembly version attribute
+            Framework = "agent_framework",
+            Runtime = "dotnet", // .NET runtime for deployment guides
+            Capabilities = new Dictionary<string, bool>
+            {
+                // Tracing capability - will be enabled when trace event support is added
+                ["tracing"] = false,
+
+                // OpenAI proxy capability - not currently supported in .NET DevUI
+                ["openai_proxy"] = false,
+
+                // Deployment capability - not currently supported in .NET DevUI
+                ["deployment"] = false
+            },
+            AuthRequired = false // Could be made configurable based on authentication middleware
+        };
+
+        return Results.Json(meta, EntitiesJsonContext.Default.MetaResponse);
+    }
+}

dotnet/src/Microsoft.Agents.AI.Hosting.OpenAI/OpenAIHostingJsonUtilities.cs

@@ -109,6 +109,7 @@ private static JsonSerializerOptions CreateDefaultOptions()
 [JsonSerializable(typeof(MCPApprovalRequestItemResource))]
 [JsonSerializable(typeof(MCPApprovalResponseItemResource))]
 [JsonSerializable(typeof(MCPCallItemResource))]
+[JsonSerializable(typeof(ExecutorActionItemResource))]
 [JsonSerializable(typeof(List<ItemResource>))]
 // ItemParam types
 [JsonSerializable(typeof(ItemParam))]

dotnet/src/Microsoft.Agents.AI.Hosting.OpenAI/Responses/AgentRunResponseUpdateExtensions.cs

@@ -45,6 +45,9 @@ public static async IAsyncEnumerable<StreamingResponseEvent> ToStreamingResponse
         var updateEnumerator = updates.GetAsyncEnumerator(cancellationToken);
         await using var _ = updateEnumerator.ConfigureAwait(false);
 
+        // Track active item IDs by executor ID to pair invoked/completed/failed events
+        Dictionary<string, string> executorItemIds = [];
+
         AgentRunResponseUpdate? previousUpdate = null;
         StreamingEventGenerator? generator = null;
         while (await updateEnumerator.MoveNextAsync().ConfigureAwait(false))
@@ -55,7 +58,92 @@ public static async IAsyncEnumerable<StreamingResponseEvent> ToStreamingResponse
             // Special-case for agent framework workflow events.
             if (update.RawRepresentation is WorkflowEvent workflowEvent)
             {
-                yield return CreateWorkflowEventResponse(workflowEvent, seq.Increment(), outputIndex);
+                // Convert executor events to standard OpenAI output_item events
+                if (workflowEvent is ExecutorInvokedEvent invokedEvent)
+                {
+                    var itemId = IdGenerator.NewId(prefix: "item");
+                    // Store the item ID for this executor so we can reuse it for completion/failure
+                    executorItemIds[invokedEvent.ExecutorId] = itemId;
+
+                    var item = new ExecutorActionItemResource
+                    {
+                        Id = itemId,
+                        ExecutorId = invokedEvent.ExecutorId,
+                        Status = "in_progress",
+                        CreatedAt = DateTimeOffset.UtcNow.ToUnixTimeSeconds()
+                    };
+
+                    yield return new StreamingOutputItemAdded
+                    {
+                        SequenceNumber = seq.Increment(),
+                        OutputIndex = outputIndex,
+                        Item = item
+                    };
+                }
+                else if (workflowEvent is ExecutorCompletedEvent completedEvent)
+                {
+                    // Reuse the item ID from the invoked event, or generate a new one if not found
+                    var itemId = executorItemIds.TryGetValue(completedEvent.ExecutorId, out var existingId)
+                        ? existingId
+                        : IdGenerator.NewId(prefix: "item");
+
+                    // Remove from tracking as this executor run is now complete
+                    executorItemIds.Remove(completedEvent.ExecutorId);
+                    JsonElement? resultData = null;
+                    if (completedEvent.Data != null && JsonSerializer.IsReflectionEnabledByDefault)
+                    {
+                        resultData = JsonSerializer.SerializeToElement(
+                            completedEvent.Data,
+                            OpenAIHostingJsonUtilities.DefaultOptions.GetTypeInfo(typeof(object)));
+                    }
+
+                    var item = new ExecutorActionItemResource
+                    {
+                        Id = itemId,
+                        ExecutorId = completedEvent.ExecutorId,
+                        Status = "completed",
+                        Result = resultData,
+                        CreatedAt = DateTimeOffset.UtcNow.ToUnixTimeSeconds()
+                    };
+
+                    yield return new StreamingOutputItemDone
+                    {
+                        SequenceNumber = seq.Increment(),
+                        OutputIndex = outputIndex,
+                        Item = item
+                    };
+                }
+                else if (workflowEvent is ExecutorFailedEvent failedEvent)
+                {
+                    // Reuse the item ID from the invoked event, or generate a new one if not found
+                    var itemId = executorItemIds.TryGetValue(failedEvent.ExecutorId, out var existingId)
+                        ? existingId
+                        : IdGenerator.NewId(prefix: "item");
+
+                    // Remove from tracking as this executor run has now failed
+                    executorItemIds.Remove(failedEvent.ExecutorId);
+
+                    var item = new ExecutorActionItemResource
+                    {
+                        Id = itemId,
+                        ExecutorId = failedEvent.ExecutorId,
+                        Status = "failed",
+                        Error = failedEvent.Data?.ToString(),
+                        CreatedAt = DateTimeOffset.UtcNow.ToUnixTimeSeconds()
+                    };
+
+                    yield return new StreamingOutputItemDone
+                    {
+                        SequenceNumber = seq.Increment(),
+                        OutputIndex = outputIndex,
+                        Item = item
+                    };
+                }
+                else
+                {
+                    // For other workflow events (not executor-specific), keep the old format as fallback
+                    yield return CreateWorkflowEventResponse(workflowEvent, seq.Increment(), outputIndex);
+                }
                 continue;
             }
 

dotnet/src/Microsoft.Agents.AI.Hosting.OpenAI/Responses/Converters/ItemResourceConverter.cs

@@ -45,6 +45,7 @@ internal sealed class ItemResourceConverter : JsonConverter<ItemResource>
             MCPApprovalRequestItemResource.ItemType => doc.Deserialize(OpenAIHostingJsonContext.Default.MCPApprovalRequestItemResource),
             MCPApprovalResponseItemResource.ItemType => doc.Deserialize(OpenAIHostingJsonContext.Default.MCPApprovalResponseItemResource),
             MCPCallItemResource.ItemType => doc.Deserialize(OpenAIHostingJsonContext.Default.MCPCallItemResource),
+            ExecutorActionItemResource.ItemType => doc.Deserialize(OpenAIHostingJsonContext.Default.ExecutorActionItemResource),
             _ => null
         };
     }
@@ -106,6 +107,9 @@ public override void Write(Utf8JsonWriter writer, ItemResource value, JsonSerial
             case MCPCallItemResource mcpCall:
                 JsonSerializer.Serialize(writer, mcpCall, OpenAIHostingJsonContext.Default.MCPCallItemResource);
                 break;
+            case ExecutorActionItemResource executorAction:
+                JsonSerializer.Serialize(writer, executorAction, OpenAIHostingJsonContext.Default.ExecutorActionItemResource);
+                break;
             default:
                 throw new JsonException($"Unknown item type: {value.GetType().Name}");
         }

dotnet/src/Microsoft.Agents.AI.Hosting.OpenAI/Responses/Models/ItemResource.cs

@@ -888,3 +888,47 @@ internal sealed class MCPCallItemResource : ItemResource
     [JsonPropertyName("error")]
     public string? Error { get; init; }
 }
+
+/// <summary>
+/// An executor action item resource for workflow execution visualization.
+/// </summary>
+internal sealed class ExecutorActionItemResource : ItemResource
+{
+    /// <summary>
+    /// The constant item type identifier for executor action items.
+    /// </summary>
+    public const string ItemType = "executor_action";
+
+    /// <inheritdoc/>
+    public override string Type => ItemType;
+
+    /// <summary>
+    /// The executor identifier.
+    /// </summary>
+    [JsonPropertyName("executor_id")]
+    public required string ExecutorId { get; init; }
+
+    /// <summary>
+    /// The execution status: "in_progress", "completed", "failed", or "cancelled".
+    /// </summary>
+    [JsonPropertyName("status")]
+    public required string Status { get; init; }
+
+    /// <summary>
+    /// The executor result data (for completed status).
+    /// </summary>
+    [JsonPropertyName("result")]
+    public JsonElement? Result { get; init; }
+
+    /// <summary>
+    /// The error message (for failed status).
+    /// </summary>
+    [JsonPropertyName("error")]
+    public string? Error { get; init; }
+
+    /// <summary>
+    /// The creation timestamp.
+    /// </summary>
+    [JsonPropertyName("created_at")]
+    public long CreatedAt { get; init; }
+}

dotnet/src/Microsoft.Agents.AI.Hosting.OpenAI/Responses/Models/StreamingResponseEvent.cs

@@ -565,7 +565,7 @@ internal sealed class StreamingWorkflowEventComplete : StreamingResponseEvent
     /// <summary>
     /// The constant event type identifier for workflow event events.
     /// </summary>
-    public const string EventType = "response.workflow_event.complete";
+    public const string EventType = "response.workflow_event.completed";
 
     /// <inheritdoc/>
     [JsonIgnore]

python/packages/devui/agent_framework_devui/_server.py

@@ -397,6 +397,7 @@ async def get_meta() -> MetaResponse:
                 ui_mode=self.mode,  # type: ignore[arg-type]
                 version=__version__,
                 framework="agent_framework",
+                runtime="python",  # Python DevUI backend
                 capabilities={
                     "tracing": os.getenv("ENABLE_OTEL") == "true",
                     "openai_proxy": openai_executor.is_configured,