Skip to content

armgit / camel

Public Imported from GitHub
0

🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

enhance prompt

Wendong748b2ecc71ce
diff --git a/camel/societies/workforce/prompts.py b/camel/societies/workforce/prompts.py
index 11aa852ae..2aaa17025 100644
--- a/camel/societies/workforce/prompts.py
+++ b/camel/societies/workforce/prompts.py
@@ -80,7 +80,7 @@ Example valid response:
]
}}
-IMPORTANT: Only add dependencies when one task truly needs the output/result of another task to complete successfully. Don't add dependencies unless they are logically necessary.
+IMPORTANT: Carefully analyze the sequence of tasks. A task's dependencies MUST include the IDs of all prior tasks whose outputs are necessary for its execution. For example, a task to 'Summarize Paper X' MUST depend on the task that 'Finds/Retrieves Paper X'. Similarly, a task that 'Compiles a report from summaries' MUST depend on all 'Summarize Paper X' tasks. Incorrect or missing dependencies will lead to critical operational failures and an inability to complete the overall objective. Be meticulous in defining these relationships.
Do not include any other text, explanations, justifications, or conversational filler before or after the JSON object. Return ONLY the JSON object.
"""
diff --git a/examples/workforce/eigent.py b/examples/workforce/eigent.py
index efbd60dd8..df4c86c0c 100644
--- a/examples/workforce/eigent.py
+++ b/examples/workforce/eigent.py
@@ -89,9 +89,9 @@ async def search_agent_factory(
human_toolkits = HumanToolkit()
tools = [
FunctionTool(search_toolkits.search_wiki),
- FunctionTool(search_toolkits.search_google),
- FunctionTool(search_toolkits.search_bing),
- FunctionTool(search_toolkits.search_baidu),
+ # FunctionTool(search_toolkits.search_google),
+ # FunctionTool(search_toolkits.search_bing),
+ # FunctionTool(search_toolkits.search_baidu),
*playwright_toolkit.get_tools(),
# *browser_toolkits.get_tools(),
*terminal_toolkits.get_tools(),
@@ -353,7 +353,7 @@ async def main():
model_backend_reason = ModelFactory.create(
model_platform=ModelPlatformType.OPENAI,
- model_type=ModelType.GPT_4_1_MINI,
+ model_type=ModelType.O3_MINI,
# model_config_dict={
# "max_tokens": 64000,
# }