refactor: Move BaseTool to main package and centralize tool description generation (#1514)

* move base_tool to main package and consolidate tool desscription generation

* update import path

* update tests

* update doc

* add base_tool test

* migrate agent delegation tools to use BaseTool

* update tests

* update import path for tool

* fix lint

* update param signature

* add from_langchain to BaseTool for backwards support of langchain tools

* fix the case where StructuredTool doesn't have func

---------

Co-authored-by: c0dez <li@vitablehealth.com>
Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>

tests/agent_test.py

@@ -5,7 +5,6 @@ from unittest import mock
 from unittest.mock import patch
 
 import pytest
-from crewai_tools import tool
 
 from crewai import Agent, Crew, Task
 from crewai.agents.cache import CacheHandler
@@ -14,6 +13,7 @@ from crewai.agents.parser import AgentAction, CrewAgentParser, OutputParserExcep
 from crewai.llm import LLM
 from crewai.tools.tool_calling import InstructorToolCalling
 from crewai.tools.tool_usage import ToolUsage
+from crewai.tools import tool
 from crewai.tools.tool_usage_events import ToolUsageFinished
 from crewai.utilities import RPMController
 from crewai.utilities.events import Emitter
@@ -277,9 +277,10 @@ def test_cache_hitting():
         "multiplier-{'first_number': 12, 'second_number': 3}": 36,
     }
 
-    with patch.object(CacheHandler, "read") as read, patch.object(
-        Emitter, "emit"
-    ) as emit:
+    with (
+        patch.object(CacheHandler, "read") as read,
+        patch.object(Emitter, "emit") as emit,
+    ):
         read.return_value = "0"
         task = Task(
             description="What is 2 times 6? Ignore correctness and just return the result of the multiplication tool, you must use the tool.",
@@ -604,7 +605,7 @@ def test_agent_respect_the_max_rpm_set(capsys):
 def test_agent_respect_the_max_rpm_set_over_crew_rpm(capsys):
     from unittest.mock import patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def get_final_answer() -> float:
@@ -642,7 +643,7 @@ def test_agent_respect_the_max_rpm_set_over_crew_rpm(capsys):
 def test_agent_without_max_rpm_respet_crew_rpm(capsys):
     from unittest.mock import patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def get_final_answer() -> float:
@@ -696,7 +697,7 @@ def test_agent_without_max_rpm_respet_crew_rpm(capsys):
 def test_agent_error_on_parsing_tool(capsys):
     from unittest.mock import patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def get_final_answer() -> float:
@@ -739,7 +740,7 @@ def test_agent_error_on_parsing_tool(capsys):
 def test_agent_remembers_output_format_after_using_tools_too_many_times():
     from unittest.mock import patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def get_final_answer() -> float:
@@ -863,11 +864,16 @@ def test_agent_function_calling_llm():
 
     from crewai.tools.tool_usage import ToolUsage
 
-    with patch.object(
-        instructor, "from_litellm", wraps=instructor.from_litellm
-    ) as mock_from_litellm, patch.object(
-        ToolUsage, "_original_tool_calling", side_effect=Exception("Forced exception")
-    ) as mock_original_tool_calling:
+    with (
+        patch.object(
+            instructor, "from_litellm", wraps=instructor.from_litellm
+        ) as mock_from_litellm,
+        patch.object(
+            ToolUsage,
+            "_original_tool_calling",
+            side_effect=Exception("Forced exception"),
+        ) as mock_original_tool_calling,
+    ):
         crew.kickoff()
         mock_from_litellm.assert_called()
         mock_original_tool_calling.assert_called()
@@ -894,7 +900,7 @@ def test_agent_count_formatting_error():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_tool_result_as_answer_is_the_final_answer_for_the_agent():
-    from crewai_tools import BaseTool
+    from crewai.tools import BaseTool
 
     class MyCustomTool(BaseTool):
         name: str = "Get Greetings"
@@ -924,7 +930,7 @@ def test_tool_result_as_answer_is_the_final_answer_for_the_agent():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_tool_usage_information_is_appended_to_agent():
-    from crewai_tools import BaseTool
+    from crewai.tools import BaseTool
 
     class MyCustomTool(BaseTool):
         name: str = "Decide Greetings"

tests/agents/agent_builder/base_agent_test.py

@@ -2,6 +2,7 @@ import hashlib
 from typing import Any, List, Optional
 
 from crewai.agents.agent_builder.base_agent import BaseAgent
+from crewai.tools.base_tool import BaseTool
 from pydantic import BaseModel
 
 
@@ -10,13 +11,13 @@ class TestAgent(BaseAgent):
         self,
         task: Any,
         context: Optional[str] = None,
-        tools: Optional[List[Any]] = None,
+        tools: Optional[List[BaseTool]] = None,
     ) -> str:
         return ""
 
     def create_agent_executor(self, tools=None) -> None: ...
 
-    def _parse_tools(self, tools: List[Any]) -> List[Any]:
+    def _parse_tools(self, tools: List[BaseTool]) -> List[BaseTool]:
         return []
 
     def get_delegation_tools(self, agents: List["BaseAgent"]): ...

tests/crew_test.py

@@ -456,7 +456,7 @@ def test_crew_verbose_output(capsys):
 def test_cache_hitting_between_agents():
     from unittest.mock import call, patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def multiplier(first_number: int, second_number: int) -> float:
@@ -499,7 +499,7 @@ def test_cache_hitting_between_agents():
 def test_api_calls_throttling(capsys):
     from unittest.mock import patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def get_final_answer() -> float:
@@ -1111,7 +1111,7 @@ def test_dont_set_agents_step_callback_if_already_set():
 def test_crew_function_calling_llm():
     from unittest.mock import patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     llm = "gpt-4o"
 
@@ -1146,7 +1146,7 @@ def test_crew_function_calling_llm():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_task_with_no_arguments():
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def return_data() -> str:
@@ -1309,8 +1309,9 @@ def test_hierarchical_crew_creation_tasks_with_agents():
 
     assert crew.manager_agent is not None
     assert crew.manager_agent.tools is not None
-    assert crew.manager_agent.tools[0].description.startswith(
-        "Delegate a specific task to one of the following coworkers: Senior Writer"
+    assert (
+        "Delegate a specific task to one of the following coworkers: Senior Writer\n"
+        in crew.manager_agent.tools[0].description
     )
 
 
@@ -1337,8 +1338,9 @@ def test_hierarchical_crew_creation_tasks_with_async_execution():
     crew.kickoff()
     assert crew.manager_agent is not None
     assert crew.manager_agent.tools is not None
-    assert crew.manager_agent.tools[0].description.startswith(
+    assert (
         "Delegate a specific task to one of the following coworkers: Senior Writer\n"
+        in crew.manager_agent.tools[0].description
     )
 
 
@@ -1370,8 +1372,9 @@ def test_hierarchical_crew_creation_tasks_with_sync_last():
     crew.kickoff()
     assert crew.manager_agent is not None
     assert crew.manager_agent.tools is not None
-    assert crew.manager_agent.tools[0].description.startswith(
+    assert (
         "Delegate a specific task to one of the following coworkers: Senior Writer, Researcher, CEO\n"
+        in crew.manager_agent.tools[0].description
     )
 
 
@@ -1494,7 +1497,7 @@ def test_task_callback_on_crew():
 def test_tools_with_custom_caching():
     from unittest.mock import patch
 
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def multiplcation_tool(first_number: int, second_number: int) -> int:
@@ -1696,7 +1699,7 @@ def test_manager_agent_in_agents_raises_exception():
 
 
 def test_manager_agent_with_tools_raises_exception():
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def testing_tool(first_number: int, second_number: int) -> int:

tests/task_test.py

@@ -15,7 +15,7 @@ from pydantic_core import ValidationError
 
 
 def test_task_tool_reflect_agent_tools():
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def fake_tool() -> None:
@@ -39,7 +39,7 @@ def test_task_tool_reflect_agent_tools():
 
 
 def test_task_tool_takes_precedence_over_agent_tools():
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def fake_tool() -> None:
@@ -656,7 +656,7 @@ def test_increment_delegations_for_sequential_process():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_increment_tool_errors():
-    from crewai_tools import tool
+    from crewai.tools import tool
 
     @tool
     def scoring_examples() -> None:

tests/tools/agent_tools/agent_tools_test.py

@@ -3,7 +3,7 @@
 import pytest
 
 from crewai.agent import Agent
-from crewai.tools.agent_tools import AgentTools
+from crewai.tools.agent_tools.agent_tools import AgentTools
 
 researcher = Agent(
     role="researcher",
@@ -11,12 +11,14 @@ researcher = Agent(
     backstory="You're an expert researcher, specialized in technology",
     allow_delegation=False,
 )
-tools = AgentTools(agents=[researcher])
+tools = AgentTools(agents=[researcher]).tools()
+delegate_tool = tools[0]
+ask_tool = tools[1]
 
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_delegate_work():
-    result = tools.delegate_work(
+    result = delegate_tool.run(
         coworker="researcher",
         task="share your take on AI Agents",
         context="I heard you hate them",
@@ -30,8 +32,8 @@ def test_delegate_work():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_delegate_work_with_wrong_co_worker_variable():
-    result = tools.delegate_work(
-        co_worker="researcher",
+    result = delegate_tool.run(
+        coworker="researcher",
         task="share your take on AI Agents",
         context="I heard you hate them",
     )
@@ -44,7 +46,7 @@ def test_delegate_work_with_wrong_co_worker_variable():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_ask_question():
-    result = tools.ask_question(
+    result = ask_tool.run(
         coworker="researcher",
         question="do you hate AI Agents?",
         context="I heard you LOVE them",
@@ -58,8 +60,8 @@ def test_ask_question():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_ask_question_with_wrong_co_worker_variable():
-    result = tools.ask_question(
-        co_worker="researcher",
+    result = ask_tool.run(
+        coworker="researcher",
         question="do you hate AI Agents?",
         context="I heard you LOVE them",
     )
@@ -72,8 +74,8 @@ def test_ask_question_with_wrong_co_worker_variable():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_delegate_work_withwith_coworker_as_array():
-    result = tools.delegate_work(
-        co_worker="[researcher]",
+    result = delegate_tool.run(
+        coworker="[researcher]",
         task="share your take on AI Agents",
         context="I heard you hate them",
     )
@@ -86,8 +88,8 @@ def test_delegate_work_withwith_coworker_as_array():
 
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_ask_question_with_coworker_as_array():
-    result = tools.ask_question(
-        co_worker="[researcher]",
+    result = ask_tool.run(
+        coworker="[researcher]",
         question="do you hate AI Agents?",
         context="I heard you LOVE them",
     )
@@ -99,7 +101,7 @@ def test_ask_question_with_coworker_as_array():
 
 
 def test_delegate_work_to_wrong_agent():
-    result = tools.ask_question(
+    result = ask_tool.run(
         coworker="writer",
         question="share your take on AI Agents",
         context="I heard you hate them",
@@ -112,7 +114,7 @@ def test_delegate_work_to_wrong_agent():
 
 
 def test_ask_question_to_wrong_agent():
-    result = tools.ask_question(
+    result = ask_tool.run(
         coworker="writer",
         question="do you hate AI Agents?",
         context="I heard you LOVE them",

tests/tools/test_base_tool.py

@@ -0,0 +1,109 @@
+from typing import Callable
+from crewai.tools import BaseTool, tool
+
+
+def test_creating_a_tool_using_annotation():
+    @tool("Name of my tool")
+    def my_tool(question: str) -> str:
+        """Clear description for what this tool is useful for, you agent will need this information to use it."""
+        return question
+
+    # Assert all the right attributes were defined
+    assert my_tool.name == "Name of my tool"
+    assert (
+        my_tool.description
+        == "Tool Name: Name of my tool\nTool Arguments: {'question': {'description': None, 'type': 'str'}}\nTool Description: Clear description for what this tool is useful for, you agent will need this information to use it."
+    )
+    assert my_tool.args_schema.schema()["properties"] == {
+        "question": {"title": "Question", "type": "string"}
+    }
+    assert (
+        my_tool.func("What is the meaning of life?") == "What is the meaning of life?"
+    )
+
+    # Assert the langchain tool conversion worked as expected
+    converted_tool = my_tool.to_langchain()
+    assert converted_tool.name == "Name of my tool"
+
+    assert (
+        converted_tool.description
+        == "Tool Name: Name of my tool\nTool Arguments: {'question': {'description': None, 'type': 'str'}}\nTool Description: Clear description for what this tool is useful for, you agent will need this information to use it."
+    )
+    assert converted_tool.args_schema.schema()["properties"] == {
+        "question": {"title": "Question", "type": "string"}
+    }
+    assert (
+        converted_tool.func("What is the meaning of life?")
+        == "What is the meaning of life?"
+    )
+
+
+def test_creating_a_tool_using_baseclass():
+    class MyCustomTool(BaseTool):
+        name: str = "Name of my tool"
+        description: str = (
+            "Clear description for what this tool is useful for, you agent will need this information to use it."
+        )
+
+        def _run(self, question: str) -> str:
+            return question
+
+    my_tool = MyCustomTool()
+    # Assert all the right attributes were defined
+    assert my_tool.name == "Name of my tool"
+
+    assert (
+        my_tool.description
+        == "Tool Name: Name of my tool\nTool Arguments: {'question': {'description': None, 'type': 'str'}}\nTool Description: Clear description for what this tool is useful for, you agent will need this information to use it."
+    )
+    assert my_tool.args_schema.schema()["properties"] == {
+        "question": {"title": "Question", "type": "string"}
+    }
+    assert my_tool.run("What is the meaning of life?") == "What is the meaning of life?"
+
+    # Assert the langchain tool conversion worked as expected
+    converted_tool = my_tool.to_langchain()
+    assert converted_tool.name == "Name of my tool"
+
+    assert (
+        converted_tool.description
+        == "Tool Name: Name of my tool\nTool Arguments: {'question': {'description': None, 'type': 'str'}}\nTool Description: Clear description for what this tool is useful for, you agent will need this information to use it."
+    )
+    assert converted_tool.args_schema.schema()["properties"] == {
+        "question": {"title": "Question", "type": "string"}
+    }
+    assert (
+        converted_tool.run("What is the meaning of life?")
+        == "What is the meaning of life?"
+    )
+
+
+def test_setting_cache_function():
+    class MyCustomTool(BaseTool):
+        name: str = "Name of my tool"
+        description: str = (
+            "Clear description for what this tool is useful for, you agent will need this information to use it."
+        )
+        cache_function: Callable = lambda: False
+
+        def _run(self, question: str) -> str:
+            return question
+
+    my_tool = MyCustomTool()
+    # Assert all the right attributes were defined
+    assert not my_tool.cache_function()
+
+
+def test_default_cache_function_is_true():
+    class MyCustomTool(BaseTool):
+        name: str = "Name of my tool"
+        description: str = (
+            "Clear description for what this tool is useful for, you agent will need this information to use it."
+        )
+
+        def _run(self, question: str) -> str:
+            return question
+
+    my_tool = MyCustomTool()
+    # Assert all the right attributes were defined
+    assert my_tool.cache_function()

tests/tools/test_tool_usage.py

@@ -3,11 +3,11 @@ import random
 from unittest.mock import MagicMock
 
 import pytest
-from crewai_tools import BaseTool
 from pydantic import BaseModel, Field
 
 from crewai import Agent, Task
 from crewai.tools.tool_usage import ToolUsage
+from crewai.tools import BaseTool
 
 
 class RandomNumberToolInput(BaseModel):
@@ -103,11 +103,7 @@ def test_tool_usage_render():
     rendered = tool_usage._render()
 
     # Updated checks to match the actual output
-    assert "Tool Name: random number generator" in rendered
-    assert (
-        "Random Number Generator(min_value: 'integer', max_value: 'integer') - Generates a random number within a specified range min_value: 'The minimum value of the range (inclusive)', max_value: 'The maximum value of the range (inclusive)'"
-        in rendered
-    )
+    assert "Tool Name: Random Number Generator" in rendered
     assert "Tool Arguments:" in rendered
     assert (
         "'min_value': {'description': 'The minimum value of the range (inclusive)', 'type': 'int'}"
@@ -117,3 +113,11 @@ def test_tool_usage_render():
         "'max_value': {'description': 'The maximum value of the range (inclusive)', 'type': 'int'}"
         in rendered
     )
+    assert (
+        "Tool Description: Generates a random number within a specified range"
+        in rendered
+    )
+    assert (
+        "Tool Name: Random Number Generator\nTool Arguments: {'min_value': {'description': 'The minimum value of the range (inclusive)', 'type': 'int'}, 'max_value': {'description': 'The maximum value of the range (inclusive)', 'type': 'int'}}\nTool Description: Generates a random number within a specified range"
+        in rendered
+    )