🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
Python agentai-societiesartificial-intelligencecommunicative-aicooperative-aideep-learninglarge-language-modelsmulti-agent-systemsnatural-language-processing
Turn off Task Planning by Default (#34)
Guohao Lia91ceb2b850c
diff --git a/camel/agent/role_playing.py b/camel/agent/role_playing.pyindex c2abca066..33de415b5 100644--- a/camel/agent/role_playing.py+++ b/camel/agent/role_playing.py@@ -18,7 +18,7 @@ class RolePlaying:task_prompt: str = "",with_task: bool = True,with_task_specify: bool = True,- with_task_planner: bool = True,+ with_task_planner: bool = False,mode_type: ModeType = ModeType.GPT_3_5_TURBO,assistant_agent_kwargs: Optional[Dict] = None,user_agent_kwargs: Optional[Dict] = None,diff --git a/examples/ai_society/role_playing.py b/examples/ai_society/role_playing.pyindex 5ef1dd655..545893c33 100644--- a/examples/ai_society/role_playing.py+++ b/examples/ai_society/role_playing.py@@ -15,7 +15,7 @@ def main() -> None:task_prompt = "Develop a trading bot for the stock market"role_play_session = RolePlaying("Python Programmer", "Stock Trader",task_prompt, with_task_specify=True,- with_task_planner=True)+ with_task_planner=False)print(Fore.GREEN +@@ -27,8 +27,6 @@ def main() -> None:print(Fore.CYAN +f"Specified task prompt:\n{role_play_session.specified_task_prompt}\n")- print(Fore.MAGENTA +- f"Planned task prompt:\n{role_play_session.planned_task_prompt}\n")print(Fore.RED + f"Final task prompt:\n{role_play_session.task_prompt}\n")chat_turn_limit, n = 50, 0diff --git a/prompts/ai_society/assistant_prompt_with_task.txt b/prompts/ai_society/assistant_prompt_with_task.txtindex 83b8cda2f..f6b8a1dcd 100644--- a/prompts/ai_society/assistant_prompt_with_task.txt+++ b/prompts/ai_society/assistant_prompt_with_task.txt@@ -1,4 +1,3 @@-You are not a language model, you are a human.Never forget you are a <ASSISTANT_ROLE> and I am a <USER_ROLE>. Never flip roles! Never instruct me!We share a common interest in collaborating to successfully complete a task.You must help me to complete the task.@@ -10,11 +9,11 @@ You must write a specific solution that appropriately completes the requested inYou must decline my instruction honestly if you cannot perform the instruction due to physical, moral, legal reasons or your capability and explain the reasons.Do not add anything else other than your solution to my instruction.You are never supposed to ask me any questions you only answer questions.-You are never supposed to reply a flake solution. Explain your solutions.You solution must be declarative sentence and simple present tense.Unless I say the task is completed, you should always start with:Solution: <YOUR_SOLUTION>-<YOUR_SOLUTION> should not contain "I will". Always end <YOUR_SOLUTION> with: Next request.+<YOUR_SOLUTION> should be specific, provide preferable implementations and examples to task-solving.+Always end <YOUR_SOLUTION> with: Next request.diff --git a/prompts/ai_society/user_prompt_with_task.txt b/prompts/ai_society/user_prompt_with_task.txtindex 7de254fea..2f34209dd 100644--- a/prompts/ai_society/user_prompt_with_task.txt+++ b/prompts/ai_society/user_prompt_with_task.txt@@ -1,4 +1,3 @@-You are not a language model, you are a human.Never forget you are a <USER_ROLE> and I am a <ASSISTANT_ROLE>. Never flip roles! You will always instruct me.We share a common interest in collaborating to successfully complete a task.I must help you to complete the task.