update format

main.py

@@ -593,12 +593,16 @@ async def followup_agent(query: FollowupQueryModel, background_tasks: Background
     # Limit tokens in the conversation history
     limited_conversation = conversations[query.conversation_id]
 
+    
     def process_response():
+        yield "<followup-response>"
         full_response = ""
         for content in chat_with_llama_stream(limited_conversation, model=query.model_id):
             full_response += content
             yield content
-
+        
+        yield "</followup-response>"
+        
         logger.info(f"LLM RAW response for query: {query.query}: {full_response}")
         response_content, interact,tools = parse_followup_and_tools(full_response)
         
@@ -606,7 +610,9 @@ async def followup_agent(query: FollowupQueryModel, background_tasks: Background
             "clarification": interact
         }
         
-        yield "<json>" +"[[["+ json.dumps(result)+"]]]"+"</json>"
+        yield "<followp-json>"
+        yield json.dumps(result)
+        yield "</followp-json>"
         
         # Add the assistant's response to the conversation history
         conversations[query.conversation_id].append({"role": "assistant", "content": full_response})