diff --git a/pilot/scene/chat_knowledge/refine_summary/prompt.py b/pilot/scene/chat_knowledge/refine_summary/prompt.py
index c2d7fb2cb..7d795b20a 100644
--- a/pilot/scene/chat_knowledge/refine_summary/prompt.py
+++ b/pilot/scene/chat_knowledge/refine_summary/prompt.py
@@ -12,11 +12,11 @@ CFG = Config()
 
 PROMPT_SCENE_DEFINE = """"""
 
-_DEFAULT_TEMPLATE_ZH = """根据提供的上下文信息，我们已经提供了一个到某一点的现有总结:{existing_answer}\n 请根据你之前推理的内容进行最终的总结,总结的时候可以详细点,回答的时候最好按照1.2.3.进行总结."""
+_DEFAULT_TEMPLATE_ZH = """根据提供的上下文信息，我们已经提供了一个到某一点的现有总结:{existing_answer}\n 请根据你之前推理的内容进行最终的总结,总结的时候需要有关键见解，并且总结回答的时候最好按照1.2.3.进行总结."""
 
 _DEFAULT_TEMPLATE_EN = """
 We have provided an existing summary up to a certain point: {existing_answer}\nWe have the opportunity to refine the existing summary (only if needed) with some more context below. 
-\nBased on the previous reasoning, please summarize the final conclusion in accordance with points 1, 2, and 3. and etc.
+\nBased on the previous reasoning, please summarize the final conclusion in accordance with points 1, 2, and 3. and give me a list of bullet points with key insights and the most important facts.
 
 """
 
diff --git a/pilot/scene/chat_knowledge/v1/chat.py b/pilot/scene/chat_knowledge/v1/chat.py
index 62301ad12..150a562da 100644
--- a/pilot/scene/chat_knowledge/v1/chat.py
+++ b/pilot/scene/chat_knowledge/v1/chat.py
@@ -49,7 +49,7 @@ class ChatKnowledge(BaseChat):
         )
         self.max_token = (
             CFG.KNOWLEDGE_SEARCH_MAX_TOKEN
-            if self.space_context is None
+            if self.space_context is None or self.space_context.get("prompt") is None
             else int(self.space_context["prompt"]["max_token"])
         )
         vector_store_config = {