clean up

byoa_tools.py

@@ -1,22 +0,0 @@
-# Initialize a ChatOpenAI model
-llm = ChatOpenAI(model="gpt-4o", temperature=0)
-
-from langgraph.prebuilt import create_react_agent
-from langgraph.checkpoint.memory import MemorySaver
-
-# Create the agent with LangGraph
-memory = MemorySaver()
-agent_executor = create_react_agent(
-    llm,
-    tools,
-    checkpointer=memory
-)
-
-# Pass the LangGraph agent to the adapter
-wrapped_agent = LangChainAgentAdapter(
-    langchain_agent=agent_executor,
-    tools=tools,
-    role="San Francisco Travel Advisor",
-    goal="Curate a detailed list of the best neighborhoods to live in, restaurants to dine at, and attractions to visit in San Francisco.",
-    backstory="An expert travel advisor with insider knowledge of San Francisco's vibrant culture, culinary delights, and hidden gems.",
-) 

src/crewai/agents/agent_builder/base_agent.py

@@ -78,6 +78,8 @@ class BaseAgent(ABC, BaseModel):
             Set the rpm controller for the agent.
         set_private_attrs() -> "BaseAgent":
             Set private attributes.
+        configure_executor(cache_handler: CacheHandler, rpm_controller: RPMController) -> None:
+            Configure the agent's executor with both cache and RPM handling.
     """
 
     __hash__ = object.__hash__  # type: ignore
@@ -163,7 +165,6 @@ class BaseAgent(ABC, BaseModel):
         tool meets these criteria, it is processed and added to the list of
         tools. Otherwise, a ValueError is raised.
         """
-        print(f"Validating tools: {tools}")
         processed_tools = []
         for tool in tools:
             if isinstance(tool, BaseTool):
@@ -181,7 +182,6 @@ class BaseAgent(ABC, BaseModel):
                     "Tool must be an instance of BaseTool or "
                     "an object with 'name', 'func', and 'description' attributes."
                 )
-        print(f"Processed tools: {processed_tools}")
         return processed_tools
 
     @model_validator(mode="after")
@@ -343,10 +343,6 @@ class BaseAgent(ABC, BaseModel):
         # Only create the executor if it hasn't been created yet.
         if self.agent_executor is None:
             self.create_agent_executor()
-        else:
-            print(
-                "Agent executor already exists, skipping creation in set_cache_handler."
-            )
 
     def increment_formatting_errors(self) -> None:
         self.formatting_errors += 1
@@ -362,7 +358,16 @@ class BaseAgent(ABC, BaseModel):
             # Only create the executor if it hasn't been created yet.
             if self.agent_executor is None:
                 self.create_agent_executor()
-            else:
+
-                print(
+    def configure_executor(
-                    "Agent executor already exists, skipping creation in set_rpm_controller."
+        self, cache_handler: CacheHandler, rpm_controller: RPMController
-                )
+    ) -> None:
+        """Configure the agent's executor with both cache and RPM handling.
+
+        This method delegates to set_cache_handler and set_rpm_controller, applying the configuration
+        only if the respective flags or values are set.
+        """
+        if self.cache:
+            self.set_cache_handler(cache_handler)
+        if self.max_rpm:
+            self.set_rpm_controller(rpm_controller)

src/crewai/agents/langchain_agent_adapter.py

@@ -161,7 +161,6 @@ class LangChainAgentAdapter(BaseAgent):
         else:
             task_prompt = self._use_trained_data(task_prompt=task_prompt)
 
-        # IMPORTANT: create an initial state using "messages" (not "input")
         init_state = {"messages": [("user", task_prompt)]}
         state = self.agent_executor.invoke(init_state)
 
@@ -176,7 +175,6 @@ class LangChainAgentAdapter(BaseAgent):
         else:
             current_output = ""
 
-        # If human feedback is required, enter a feedback loop
         if task.human_input:
             current_output = self._handle_human_feedback(current_output)
 
@@ -206,7 +204,6 @@ class LangChainAgentAdapter(BaseAgent):
                 f"Updated answer:"
             )
             try:
-                # Use "messages" key for the prompt, like we do in execute_task.
                 new_state = self.agent_executor.invoke(
                     {"messages": [("user", new_prompt)]}
                 )
@@ -309,9 +306,6 @@ class LangChainAgentAdapter(BaseAgent):
             else:
                 used_tools.append(tool)
 
-        print("Raw tools:", raw_tools)
-        print("Used tools:", used_tools)
-
         # Sanitize the agent's role for the "name" field. The allowed pattern is ^[a-zA-Z0-9_-]+$
         import re
 

src/crewai/crew.py

@@ -340,10 +340,7 @@ class Crew(BaseModel):
 
         if self.agents:
             for agent in self.agents:
-                if self.cache:
+                agent.configure_executor(self._cache_handler, self._rpm_controller)
-                    agent.set_cache_handler(self._cache_handler)
-                if self.max_rpm:
-                    agent.set_rpm_controller(self._rpm_controller)
         return self
 
     @model_validator(mode="after")

src/crewai/tools/base_tool.py

@@ -92,7 +92,6 @@ class BaseTool(BaseModel, ABC):
 
     @classmethod
     def from_langchain(cls, tool: Any) -> "BaseTool":
-        print("CREATING TOOL FROM LANGCHAIN")
         """Create a Tool instance from a CrewStructuredTool.
 
         This method takes a CrewStructuredTool object and converts it into a