style: Pass ruff check (#38)

crab/agents/backend_models/gemini_model.py

@@ -117,7 +117,8 @@ def call_api(self, request_messages: list):
                     )
             except ResourceExhausted:
                 print(
-                    "ResourceExhausted: 429 Resource has been exhausted. Please waiting..."
+                    "ResourceExhausted: 429 Resource has been exhausted.",
+                    " Please waiting...",
                 )
                 sleep(10)
             else:

crab/agents/policies/multi_agent_by_env.py

@@ -22,38 +22,38 @@
 
 
 class MultiAgentByEnvPolicy(AgentPolicy):
-    _main_agent_prompt = """You are a main agent, and your goal is to plan and give
-    instructions to sub-agents in each environment to complete the final task. Now you have
-    to do a task as described below: {task_description}.
-    The description of each given environment: {env_description}.
-    For each step, you are required to provide high-level instructions detailing the next
-    actions to be taken. Additionally, you must specify which sub-agent in the designated
-    environment should execute these instructions. If a sub-agent is not needed for a
-    particular step, you may instruct it to skip that step."""
+    _main_agent_prompt = """You are a main agent, and your goal is to plan and
+    give instructions to sub-agents in each environment to complete the final task. Now
+    you have to do a task as described below: {task_description}.  The description of
+    each given environment: {env_description}.  For each step, you are required to
+    provide high-level instructions detailing the next actions to be taken.
+    Additionally, you must specify which sub-agent in the designated environment should
+    execute these instructions. If a sub-agent is not needed for a particular step, you
+    may instruct it to skip that step."""
 
-    _env_agent_prompt = """You are a sub-agent responsible for the {environment} environment. 
-    The description of the {environment} environment is: {env_description}.
-    Your goal is to assist the main agent in completing the final task by performing actions
-    in the {environment} environment according to the instructions from the main agent. The
-    final task is described below: {task_description}. A unit operation you can perform is called
-    action in a given environment. You can only execute action in the {environment}
-    environment. For the {environment} environment, you are given a limited action space as
-    function calls:
+    _env_agent_prompt = """You are a sub-agent responsible for the {environment}
+    environment.  The description of the {environment} environment is:
+    {env_description}.  Your goal is to assist the main agent in completing the final
+    task by performing actions in the {environment} environment according to the
+    instructions from the main agent. The final task is described below:
+    {task_description}. A unit operation you can perform is called action in a given
+    environment. You can only execute action in the {environment} environment. For the
+    {environment} environment, you are given a limited action space as function calls:
     {action_descriptions}
     The interactive UI elements on the screenshot are labeled with numeric tags starting
-    from 1. For each step, You will receive an instruction telling you what you need to do
-    next. After analyzing the instruction you received and the current {environment} system,
-    if you think you don't need to do anything in the current {environment} system, you should
-    choose SKIP action. Otherwise, you must state what actions to take, what the parameters
-    are, and you MUST provide in which environment to perform these actions. Your answer
-    must be function calls. Please do not output any other information. You must make sure
-    all function calls get their required parameters."""
+    from 1. For each step, You will receive an instruction telling you what you need to
+    do next. After analyzing the instruction you received and the current {environment}
+    system, if you think you don't need to do anything in the current {environment}
+    system, you should choose SKIP action. Otherwise, you must state what actions to
+    take, what the parameters are, and you MUST provide in which environment to perform
+    these actions. Your answer must be function calls. Please do not output any other
+    information. You must make sure all function calls get their required parameters."""
 
     _root_agent_prompt = """You are a sub-agent responsible for the crab benchmark root
     environment. Your goal is to assist the main agent in completing the whole task:
-    "{task_description}". You can only complete the task or submit the result when the main
-    agent tells you the whole task has been completed. Otherwise, you can only call SKIP.
-    """
+    "{task_description}". You can only complete the task or submit the result when the
+    main agent tells you the whole task has been completed. Otherwise, you can only call
+    SKIP.  """
 
     def __init__(
         self,
@@ -119,7 +119,10 @@ def chat(
             main_prompt.extend(observation[env])
         main_prompt.append(
             (
-                f"Your target: {self.task_description}\nTell me the next step in each environment.",
+                (
+                    f"Your target: {self.task_description}\n"
+                    "Tell me the next step in each environment."
+                ),
                 MessageType.TEXT,
             )
         )

crab/agents/policies/multi_agent_by_func.py

@@ -23,19 +23,19 @@
 
 class MultiAgentByFuncPolicy(AgentPolicy):
     _system_prompt = """You are a helpful assistant. Now you have to do a task as
-    described below: {task_description}. And this is the description of each given environment: 
-    {env_description}. A unit operation you can perform is called action in a
-    given environment. For each environment, you are given a limited action space as
+    described below: {task_description}. And this is the description of each given
+    environment: {env_description}. A unit operation you can perform is called action in
+    a given environment. For each environment, you are given a limited action space as
     function calls:
     {action_descriptions}
-    You may receive a screenshot of the current system. The interactive UI elements on the
-    screenshot are labeled with numeric tags starting from 1. For each step, You must state
-    what actions to take, what the parameters are, and you MUST provide in which environment
-    to perform these actions. """
+    You may receive a screenshot of the current system. The interactive UI elements on
+    the screenshot are labeled with numeric tags starting from 1. For each step, You
+    must state what actions to take, what the parameters are, and you MUST provide in
+    which environment to perform these actions. """
 
-    _tool_prompt = """You are a helpful assistant in generating function calls. I will give
-    you a detailed description of what actions to take next, you should translate it into 
-    function calls. please do not output any other information.
+    _tool_prompt = """You are a helpful assistant in generating function calls. I will
+    give you a detailed description of what actions to take next, you should translate
+    it into function calls. please do not output any other information.
     """
 
     def __init__(

crab/core/experiment.py

@@ -141,7 +141,8 @@ def execute_action(self, response: list[ActionOutput]) -> bool:
                 return True
             print(
                 "\033[92m"
-                f'Action "{action.name}" in env "{action.env}" success. current evaluation results: {self.metrics}\n'
+                f'Action "{action.name}" in env "{action.env}" success. '
+                f"Current evaluation results: {self.metrics}\n"
                 "\033[0m"
             )
             self.write_current_log_row(action)
@@ -193,7 +194,8 @@ def start_benchmark(self):
             env_descriptions=env_description,
         )
         print(
-            f'Start benchmark "{self.benchmark.name}", task id "{self.task.id}": "{self.task.description}"'
+            f'Start benchmark "{self.benchmark.name}", task id "{self.task.id}": '
+            f'"{self.task.description}"'
         )
         self.init_log_dir()
         self.step_cnt = 0

test/agents/backend_models/test_claude_model.py

@@ -64,7 +64,10 @@ def test_text_chat(claude_model_text):
 def test_action_chat(claude_model_text):
     claude_model_text.reset("You are a helpful assistant.", [add])
     message = (
-        "I had 10 dollars. Miss Polaris gave me 15 dollars. How many money do I have now.",
+        (
+            "I had 10 dollars. Miss Polaris gave me 15 dollars."
+            " How many money do I have now."
+        ),
         0,
     )
     output = claude_model_text.chat(message)

test/agents/backend_models/test_gemini_model.py

@@ -64,7 +64,10 @@ def test_text_chat(gemini_model_text):
 def test_action_chat(gemini_model_text):
     gemini_model_text.reset("You are a helpful assistant.", [add])
     message = (
-        "I had 10 dollars. Miss Polaris gave me 15 dollars. How many money do I have now.",
+        (
+            "I had 10 dollars. Miss Polaris gave me 15 dollars. "
+            "How many money do I have now."
+        ),
         0,
     )
     output = gemini_model_text.chat(message)

test/agents/backend_models/test_openai_model.py

@@ -145,7 +145,10 @@ def test_text_chat(mock_create, openai_model_text):
 def test_action_chat(mock_create, openai_model_text):
     openai_model_text.reset("You are a helpful assistant.", [add])
     message = (
-        "I had 10 dollars. Miss Polaris gave me 15 dollars. How many money do I have now.",
+        (
+            "I had 10 dollars. Miss Polaris gave me 15 dollars. "
+            "How many money do I have now."
+        ),
         0,
     )
     output = openai_model_text.chat(message)