Merge pull request #85 from alipay/dev_chongshi_2

Improvement: tweak comments in the prompt_util file

agentuniverse/base/util/prompt_util.py

@@ -164,8 +164,8 @@ def process_llm_token(agent_llm: LLM, lc_prompt_template, profile: dict, planner
 
     input_tokens = agent_llm.max_context_length() - agent_llm.max_tokens
     if input_tokens <= 0:
-        raise Exception("The `max_tokens` in the llm configuration is the maximum output number of tokens, "
-                        "the current `max_tokens` is greater than the context length of the LLM model.")
+        raise Exception("The current output max tokens limit is greater than the context length of the LLM model, "
+                        "please adjust it by editing the `max_tokens` parameter in the llm yaml.")
 
     if prompt_tokens <= input_tokens:
         return