Display actions in frontend (needs to add manifest files to LibreChat)

app.py

@@ -170,19 +170,19 @@ async def chat_completions(chat_request: ChatCompletionRequest):
         f"Chat request - stream={chat_request.stream}, input_length={len(user_input)}"
     )
 
-    try:
-        answer = agent.step(user_input)
-    except LLMAPIError as e:
-        logger.error(f"LLM API error: {e}")
-        raise HTTPException(status_code=502, detail=f"LLM API error: {e}")
-    except Exception as e:
-        logger.error(f"Agent error: {e}", exc_info=True)
-        raise HTTPException(status_code=500, detail="Internal agent error")
-
     created_ts = int(time.time())
     completion_id = f"chatcmpl-{uuid.uuid4().hex}"
 
     if not chat_request.stream:
+        try:
+            answer = agent.step(user_input)
+        except LLMAPIError as e:
+            logger.error(f"LLM API error: {e}")
+            raise HTTPException(status_code=502, detail=f"LLM API error: {e}")
+        except Exception as e:
+            logger.error(f"Agent error: {e}", exc_info=True)
+            raise HTTPException(status_code=500, detail="Internal agent error")
+
         return JSONResponse(
             {
                 "id": completion_id,
@@ -205,20 +205,44 @@ async def chat_completions(chat_request: ChatCompletionRequest):
         )
 
     async def event_generator():
-        chunk = {
-            "id": completion_id,
-            "object": "chat.completion.chunk",
-            "created": created_ts,
-            "model": chat_request.model,
-            "choices": [
-                {
-                    "index": 0,
-                    "delta": {"role": "assistant", "content": answer or ""},
-                    "finish_reason": "stop",
-                }
-            ],
-        }
-        yield f"data: {json.dumps(chunk, ensure_ascii=False)}\n\n"
-        yield "data: [DONE]\n\n"
+        try:
+            # Stream the agent execution
+            async for chunk in agent.step_streaming(user_input, completion_id, created_ts, chat_request.model):
+                yield f"data: {json.dumps(chunk, ensure_ascii=False)}\n\n"
+            yield "data: [DONE]\n\n"
+        except LLMAPIError as e:
+            logger.error(f"LLM API error: {e}")
+            error_chunk = {
+                "id": completion_id,
+                "object": "chat.completion.chunk",
+                "created": created_ts,
+                "model": chat_request.model,
+                "choices": [
+                    {
+                        "index": 0,
+                        "delta": {"role": "assistant", "content": f"Error: {e}"},
+                        "finish_reason": "stop",
+                    }
+                ],
+            }
+            yield f"data: {json.dumps(error_chunk, ensure_ascii=False)}\n\n"
+            yield "data: [DONE]\n\n"
+        except Exception as e:
+            logger.error(f"Agent error: {e}", exc_info=True)
+            error_chunk = {
+                "id": completion_id,
+                "object": "chat.completion.chunk",
+                "created": created_ts,
+                "model": chat_request.model,
+                "choices": [
+                    {
+                        "index": 0,
+                        "delta": {"role": "assistant", "content": "Internal agent error"},
+                        "finish_reason": "stop",
+                    }
+                ],
+            }
+            yield f"data: {json.dumps(error_chunk, ensure_ascii=False)}\n\n"
+            yield "data: [DONE]\n\n"
 
     return StreamingResponse(event_generator(), media_type="text/event-stream")