fix: Fix planning_llm issue (#1189)

* fix: Fix planning_llm issue

* fix: add poetry.lock updated version

* fix: type checking issues

* fix: tests

src/crewai/crew.py

@@ -1,9 +1,9 @@
 import asyncio
 import json
+import os
 import uuid
 from concurrent.futures import Future
 from hashlib import md5
-import os
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Union
 
 from langchain_core.callbacks import BaseCallbackHandler
@@ -48,11 +48,10 @@
 from crewai.utilities.task_output_storage_handler import TaskOutputStorageHandler
 from crewai.utilities.training_handler import CrewTrainingHandler
 
-
 agentops = None
 if os.environ.get("AGENTOPS_API_KEY"):
     try:
-        import agentops
+        import agentops  # type: ignore
     except ImportError:
         pass
 
@@ -541,7 +540,7 @@ def _handle_crew_planning(self):
         )._handle_crew_planning()
 
         for task, step_plan in zip(self.tasks, result.list_of_plans_per_task):
-            task.description += step_plan
+            task.description += step_plan.plan
 
     def _store_execution_log(
         self,

src/crewai/telemetry/telemetry.py

@@ -295,7 +295,7 @@ def tool_usage_error(self, llm: Any):
                 pass
 
     def individual_test_result_span(
-        self, crew: Crew, quality: int, exec_time: int, model_name: str
+        self, crew: Crew, quality: float, exec_time: int, model_name: str
     ):
         if self.ready:
             try:

src/crewai/tools/tool_usage.py

@@ -1,6 +1,6 @@
 import ast
-from difflib import SequenceMatcher
 import os
+from difflib import SequenceMatcher
 from textwrap import dedent
 from typing import Any, List, Union
 
@@ -15,7 +15,7 @@
 agentops = None
 if os.environ.get("AGENTOPS_API_KEY"):
     try:
-        import agentops
+        import agentops  # type: ignore
     except ImportError:
         pass
 
@@ -71,14 +71,14 @@ def __init__(
         self.task = task
         self.action = action
         self.function_calling_llm = function_calling_llm
-      
+
         # Handling bug (see https://github.com/langchain-ai/langchain/pull/16395): raise an error if tools_names have space for ChatOpenAI
         if isinstance(self.function_calling_llm, ChatOpenAI):
             if " " in self.tools_names:
                 raise Exception(
                     "Tools names should not have spaces for ChatOpenAI models."
                 )
-              
+
         # Set the maximum parsing attempts for bigger models
         if (isinstance(self.function_calling_llm, ChatOpenAI)) and (
             self.function_calling_llm.openai_api_base is None
@@ -118,7 +118,7 @@ def _use(
         tool: BaseTool,
         calling: Union[ToolCalling, InstructorToolCalling],
     ) -> str:  # TODO: Fix this return type
-        tool_event = agentops.ToolEvent(name=calling.tool_name) if agentops else None
+        tool_event = agentops.ToolEvent(name=calling.tool_name) if agentops else None  # type: ignore
         if self._check_tool_repeated_usage(calling=calling):  # type: ignore # _check_tool_repeated_usage of "ToolUsage" does not return a value (it only ever returns None)
             try:
                 result = self._i18n.errors("task_repeated_usage").format(

src/crewai/utilities/planning_handler.py

@@ -1,14 +1,25 @@
 from typing import Any, List, Optional
 
 from langchain_openai import ChatOpenAI
-from pydantic import BaseModel
+from pydantic import BaseModel, Field
 
 from crewai.agent import Agent
 from crewai.task import Task
 
 
+class PlanPerTask(BaseModel):
+    task: str = Field(..., description="The task for which the plan is created")
+    plan: str = Field(
+        ...,
+        description="The step by step plan on how the agents can execute their tasks using the available tools with mastery",
+    )
+
+
 class PlannerTaskPydanticOutput(BaseModel):
-    list_of_plans_per_task: List[str]
+    list_of_plans_per_task: List[PlanPerTask] = Field(
+        ...,
+        description="Step by step plan on how the agents can execute their tasks using the available tools with mastery",
+    )
 
 
 class CrewPlanner:

tests/utilities/test_planning_handler.py

@@ -6,7 +6,11 @@
 from crewai.agent import Agent
 from crewai.task import Task
 from crewai.tasks.task_output import TaskOutput
-from crewai.utilities.planning_handler import CrewPlanner, PlannerTaskPydanticOutput
+from crewai.utilities.planning_handler import (
+    CrewPlanner,
+    PlannerTaskPydanticOutput,
+    PlanPerTask,
+)
 
 
 class TestCrewPlanner:
@@ -44,12 +48,17 @@ def crew_planner_different_llm(self):
         return CrewPlanner(tasks, planning_agent_llm)
 
     def test_handle_crew_planning(self, crew_planner):
+        list_of_plans_per_task = [
+            PlanPerTask(task="Task1", plan="Plan 1"),
+            PlanPerTask(task="Task2", plan="Plan 2"),
+            PlanPerTask(task="Task3", plan="Plan 3"),
+        ]
         with patch.object(Task, "execute_sync") as execute:
             execute.return_value = TaskOutput(
                 description="Description",
                 agent="agent",
                 pydantic=PlannerTaskPydanticOutput(
-                    list_of_plans_per_task=["Plan 1", "Plan 2", "Plan 3"]
+                    list_of_plans_per_task=list_of_plans_per_task
                 ),
             )
             result = crew_planner._handle_crew_planning()
@@ -91,7 +100,9 @@ def test_handle_crew_planning_different_llm(self, crew_planner_different_llm):
             execute.return_value = TaskOutput(
                 description="Description",
                 agent="agent",
-                pydantic=PlannerTaskPydanticOutput(list_of_plans_per_task=["Plan 1"]),
+                pydantic=PlannerTaskPydanticOutput(
+                    list_of_plans_per_task=[PlanPerTask(task="Task1", plan="Plan 1")]
+                ),
             )
             result = crew_planner_different_llm._handle_crew_planning()