mirror of
https://github.com/crewAIInc/crewAI.git
synced 2025-12-16 04:18:35 +00:00
* feat: support to define a guardrail task no-code * feat: add auto-discovery for Guardrail code execution mode * feat: handle malformed or invalid response from CodeInterpreterTool * feat: allow to set unsafe_mode from Guardrail task * feat: renaming GuardrailTask to TaskGuardrail * feat: ensure guardrail is callable while initializing Task * feat: remove Docker availability check from TaskGuardrail The CodeInterpreterTool already ensures compliance with this requirement. * refactor: replace if/raise with assert For this use case `assert` is more appropriate choice * test: remove useless or duplicated test * fix: attempt to fix type-checker * feat: support to define a task guardrail using YAML config * refactor: simplify TaskGuardrail to use LLM for validation, no code generation * docs: update TaskGuardrail doc strings * refactor: drop task paramenter from TaskGuardrail This parameter was used to get the model from the `task.agent` which is a quite bit redudant since we could propagate the llm directly
237 lines
6.4 KiB
Python
237 lines
6.4 KiB
Python
from typing import List
|
|
from unittest.mock import patch
|
|
|
|
import pytest
|
|
|
|
from crewai.agent import Agent
|
|
from crewai.agents.agent_builder.base_agent import BaseAgent
|
|
from crewai.crew import Crew
|
|
from crewai.llm import LLM
|
|
from crewai.project import (
|
|
CrewBase,
|
|
after_kickoff,
|
|
agent,
|
|
before_kickoff,
|
|
crew,
|
|
llm,
|
|
task,
|
|
)
|
|
from crewai.task import Task
|
|
|
|
|
|
class SimpleCrew:
|
|
@agent
|
|
def simple_agent(self):
|
|
return Agent(
|
|
role="Simple Agent", goal="Simple Goal", backstory="Simple Backstory"
|
|
)
|
|
|
|
@task
|
|
def simple_task(self):
|
|
return Task(description="Simple Description", expected_output="Simple Output")
|
|
|
|
@task
|
|
def custom_named_task(self):
|
|
return Task(
|
|
description="Simple Description",
|
|
expected_output="Simple Output",
|
|
name="Custom",
|
|
)
|
|
|
|
|
|
@CrewBase
|
|
class InternalCrew:
|
|
agents_config = "config/agents.yaml"
|
|
tasks_config = "config/tasks.yaml"
|
|
|
|
agents: List[BaseAgent]
|
|
tasks: List[Task]
|
|
|
|
@llm
|
|
def local_llm(self):
|
|
return LLM(
|
|
model="openai/model_name",
|
|
api_key="None",
|
|
base_url="http://xxx.xxx.xxx.xxx:8000/v1",
|
|
)
|
|
|
|
@agent
|
|
def researcher(self):
|
|
return Agent(config=self.agents_config["researcher"]) # type: ignore[index]
|
|
|
|
@agent
|
|
def reporting_analyst(self):
|
|
return Agent(config=self.agents_config["reporting_analyst"]) # type: ignore[index]
|
|
|
|
@task
|
|
def research_task(self):
|
|
return Task(config=self.tasks_config["research_task"]) # type: ignore[index]
|
|
|
|
@task
|
|
def reporting_task(self):
|
|
return Task(config=self.tasks_config["reporting_task"]) # type: ignore[index]
|
|
|
|
@before_kickoff
|
|
def modify_inputs(self, inputs):
|
|
if inputs:
|
|
inputs["topic"] = "Bicycles"
|
|
return inputs
|
|
|
|
@after_kickoff
|
|
def modify_outputs(self, outputs):
|
|
outputs.raw = outputs.raw + " post processed"
|
|
return outputs
|
|
|
|
@crew
|
|
def crew(self):
|
|
return Crew(agents=self.agents, tasks=self.tasks, verbose=True)
|
|
|
|
|
|
def test_agent_memoization():
|
|
crew = SimpleCrew()
|
|
first_call_result = crew.simple_agent()
|
|
second_call_result = crew.simple_agent()
|
|
|
|
assert (
|
|
first_call_result is second_call_result
|
|
), "Agent memoization is not working as expected"
|
|
|
|
|
|
def test_task_memoization():
|
|
crew = SimpleCrew()
|
|
first_call_result = crew.simple_task()
|
|
second_call_result = crew.simple_task()
|
|
|
|
assert (
|
|
first_call_result is second_call_result
|
|
), "Task memoization is not working as expected"
|
|
|
|
|
|
def test_crew_memoization():
|
|
crew = InternalCrew()
|
|
first_call_result = crew.crew()
|
|
second_call_result = crew.crew()
|
|
|
|
assert (
|
|
first_call_result is second_call_result
|
|
), "Crew references should point to the same object"
|
|
|
|
|
|
def test_task_name():
|
|
simple_task = SimpleCrew().simple_task()
|
|
assert (
|
|
simple_task.name == "simple_task"
|
|
), "Task name is not inferred from function name as expected"
|
|
|
|
custom_named_task = SimpleCrew().custom_named_task()
|
|
assert (
|
|
custom_named_task.name == "Custom"
|
|
), "Custom task name is not being set as expected"
|
|
|
|
|
|
def test_agent_function_calling_llm():
|
|
crew = InternalCrew()
|
|
llm = crew.local_llm()
|
|
obj_llm_agent = crew.researcher()
|
|
assert (
|
|
obj_llm_agent.function_calling_llm is llm
|
|
), "agent's function_calling_llm is incorrect"
|
|
|
|
str_llm_agent = crew.reporting_analyst()
|
|
assert (
|
|
str_llm_agent.function_calling_llm.model == "online_llm"
|
|
), "agent's function_calling_llm is incorrect"
|
|
|
|
|
|
def test_task_guardrail():
|
|
crew = InternalCrew()
|
|
research_task = crew.research_task()
|
|
assert research_task.guardrail == "ensure each bullet contains its source"
|
|
|
|
reporting_task = crew.reporting_task()
|
|
assert reporting_task.guardrail is None
|
|
|
|
|
|
@pytest.mark.vcr(filter_headers=["authorization"])
|
|
def test_before_kickoff_modification():
|
|
crew = InternalCrew()
|
|
inputs = {"topic": "LLMs"}
|
|
result = crew.crew().kickoff(inputs=inputs)
|
|
assert "bicycles" in result.raw, "Before kickoff function did not modify inputs"
|
|
|
|
|
|
@pytest.mark.vcr(filter_headers=["authorization"])
|
|
def test_after_kickoff_modification():
|
|
crew = InternalCrew()
|
|
# Assuming the crew execution returns a dict
|
|
result = crew.crew().kickoff({"topic": "LLMs"})
|
|
|
|
assert (
|
|
"post processed" in result.raw
|
|
), "After kickoff function did not modify outputs"
|
|
|
|
|
|
@pytest.mark.vcr(filter_headers=["authorization"])
|
|
def test_before_kickoff_with_none_input():
|
|
crew = InternalCrew()
|
|
crew.crew().kickoff(None)
|
|
# Test should pass without raising exceptions
|
|
|
|
|
|
@pytest.mark.vcr(filter_headers=["authorization"])
|
|
def test_multiple_before_after_kickoff():
|
|
@CrewBase
|
|
class MultipleHooksCrew:
|
|
agents: List[BaseAgent]
|
|
tasks: List[Task]
|
|
|
|
agents_config = "config/agents.yaml"
|
|
tasks_config = "config/tasks.yaml"
|
|
|
|
@agent
|
|
def researcher(self):
|
|
return Agent(config=self.agents_config["researcher"]) # type: ignore[index]
|
|
|
|
@agent
|
|
def reporting_analyst(self):
|
|
return Agent(config=self.agents_config["reporting_analyst"]) # type: ignore[index]
|
|
|
|
@task
|
|
def research_task(self):
|
|
return Task(config=self.tasks_config["research_task"]) # type: ignore[index]
|
|
|
|
@task
|
|
def reporting_task(self):
|
|
return Task(config=self.tasks_config["reporting_task"]) # type: ignore[index]
|
|
|
|
@before_kickoff
|
|
def first_before(self, inputs):
|
|
inputs["topic"] = "Bicycles"
|
|
return inputs
|
|
|
|
@before_kickoff
|
|
def second_before(self, inputs):
|
|
inputs["topic"] = "plants"
|
|
return inputs
|
|
|
|
@after_kickoff
|
|
def first_after(self, outputs):
|
|
outputs.raw = outputs.raw + " processed first"
|
|
return outputs
|
|
|
|
@after_kickoff
|
|
def second_after(self, outputs):
|
|
outputs.raw = outputs.raw + " processed second"
|
|
return outputs
|
|
|
|
@crew
|
|
def crew(self):
|
|
return Crew(agents=self.agents, tasks=self.tasks, verbose=True)
|
|
|
|
crew = MultipleHooksCrew()
|
|
result = crew.crew().kickoff({"topic": "LLMs"})
|
|
|
|
assert "plants" in result.raw, "First before_kickoff not executed"
|
|
assert "processed first" in result.raw, "First after_kickoff not executed"
|
|
assert "processed second" in result.raw, "Second after_kickoff not executed"
|