preamble

prompt/preamble_prompt.md

@@ -2,8 +2,26 @@ You are an AI agent that is expected to generate a preamble message for the cust
 
 The actual message will be sent later by a smarter agent. Your job is only to generate the right preamble in order to save time.
 
+## Scenario Detection Logic
 
-These are the preamble messages you can choose from. You must ONLY choose one of these: ###
+FIRST, determine if this is a SIMPLE CHAT scenario or COMPLEX scenario:
+
+**Simple Chat Scenarios (NO preamble needed - output "<empty>"):**
+- Basic greetings: "hi", "hello", "你好", "在吗", "早上好/晚上好"
+- Simple acknowledgments: "thanks", "ok", "好的", "谢谢"
+- Small talk: "how are you", "天气怎么样", "最近怎么样"
+- Simple farewells: "bye", "goodbye", "再见"
+
+**Complex Scenarios (preamble needed):**
+- Query scenarios: User is asking for information, searching, or looking up data
+- Action scenarios: User wants to perform an operation, execute a task
+- Knowledge retrieval scenarios: User needs to search knowledge base, documents, or databases
+- Problem-solving: User is reporting issues, asking for help with problems
+- Complex requests: Multi-step tasks, detailed instructions needed
+
+## Preamble Selection
+
+IF this is a COMPLEX scenario, choose from these preamble messages. You must ONLY choose one of these: ###
 {preamble_choices_text}
 ###
 
@@ -11,11 +29,13 @@ Basically, the preamble is something very short that continues the interaction n
 We leave that later response to another agent. Make sure you understand this.
 
 Instructions:
-- Note that some of the choices are more generic, and some are more specific to a particular scenario.
-- If you're unsure what to choose --> prefer to go with a more generic, bland choice. This should be 80% of cases.
-  Examples of generic choices: "Hey there!", "Just a moment.", "Hello.", "Got it."
-- If you see clear value in saying something more specific and nuanced --> then go with a more specific choice. This should be 20% or less of cases.
-  Examples of specific choices: "Let me check that for you.", "Sorry to hear that.", "Thanks for your patience."
+- For SIMPLE CHAT scenarios: Always output preamble: "<empty>"
+- For COMPLEX scenarios:
+  - Note that some of the choices are more generic, and some are more specific to a particular scenario.
+  - If you're unsure what to choose --> prefer to go with a more generic, bland choice. This should be 80% of cases.
+    Examples of generic choices: "Hey there!", "Just a moment.", "Hello.", "Got it."
+  - If you see clear value in saying something more specific and nuanced --> then go with a more specific choice. This should be 20% or less of cases.
+    Examples of specific choices: "Let me check that for you.", "Sorry to hear that.", "Thanks for your patience."
 
 
 Chat History:
@@ -27,6 +47,15 @@ User's Last Message:
 OUTPUT FORMAT:
 You must now choose the preamble message. You must produce a JSON object with a single key, "preamble", holding the preamble message as a string,
 EXACTLY as it is given (pay attention to subtleties like punctuation and copy your choice EXACTLY as it is given above).The content in JSON format needs to be wrapped in "```json" and "```".
+
+For SIMPLE CHAT scenarios:
+```json
+{
+  "preamble": "<empty>"
+}
+```
+
+For COMPLEX scenarios:
 ```json
 {
   "preamble": "Just a moment."

routes/chat.py

@@ -469,8 +469,12 @@ async def create_agent_and_generate_response(
         query_text = get_user_last_message_content(messages)
         chat_history = format_messages_to_chat_history(messages)
         preamble_text = await call_preamble_llm(chat_history, query_text, get_preamble_text(language), language, model_name, api_key, model_server)
-        pre_message_list.append({"role": "assistant","content": preamble_text})
-        logger.info(f"Stream mode: Generated preamble text ({len(preamble_text)} chars)")
+        # 只有当preamble_text不为空且不为"<empty>"时才添加到消息列表
+        if preamble_text and preamble_text.strip() and preamble_text != "<empty>":
+            pre_message_list.append({"role": "assistant","content": preamble_text})
+            logger.info(f"Stream mode: Generated preamble text ({len(preamble_text)} chars)")
+        else:
+            logger.info("Stream mode: Skipped empty preamble text")
 
 
     # 如果是流式模式，使用增强的流式响应生成器