Python: [Bug]: `STORES_BY_DEFAULT` seems to cause agents have neither local nor remote storage

[TaoChenOSU]

Description

Just discovered a new flag that controls whether chat history is stored locally or remotely: STORE_BY_DEFAULT. Some clients have this set to True: https://github.com/microsoft/agent-framework/blob/main/python/packages/core/agent_framework/openai/_responses_client.py#L250

However, it looks like there is a bug where if store=False is passed the agent that uses one of these clients, neither a local nor a remote storage is set up for the agent: https://github.com/microsoft/agent-framework/blob/main/python/packages/core/agent_framework/_agents.py#L1041

Code Sample

# Copyright (c) Microsoft. All rights reserved.

import asyncio
import os
from typing import Annotated, Any

from agent_framework import AgentSession, Message, tool
from agent_framework.azure import AzureOpenAIResponsesClient
from azure.identity import AzureCliCredential
from dotenv import load_dotenv
from pydantic import Field

# Load environment variables from .env file
load_dotenv()


@tool(approval_mode="always_require")
async def get_weather(
    location: Annotated[str, Field(description="The location to get the weather for.")],
    **kwargs: Any,
) -> str:
    """Get the weather for a given location."""
    # Get session object from kwargs
    session = kwargs.get("session")
    if session and isinstance(session, AgentSession) and session.service_session_id:
        print(f"Session ID: {session.service_session_id}.")

    return f"The weather in {location} is cloudy."


async def main() -> None:
    client = AzureOpenAIResponsesClient(
        project_endpoint=os.environ["AZURE_AI_PROJECT_ENDPOINT"],
        deployment_name=os.environ["AZURE_AI_MODEL_DEPLOYMENT_NAME"],
        credential=AzureCliCredential(),
    )
    agent = client.as_agent(
        name="WeatherAgent",
        instructions="You are a helpful weather assistant.",
        tools=[get_weather],
        default_options={"store": False},
    )

    session = agent.create_session()
    result = await agent.run("What is the weather in London?", session=session)
    while len(result.user_input_requests) > 0:
        approvals: list[Message] = []
        for user_input_needed in result.user_input_requests:
            print(
                f"\nUser Input Request for function from {agent.name}:"
                f"\n  Function: {user_input_needed.function_call.name}"
                f"\n  Arguments: {user_input_needed.function_call.arguments}"
            )

            # Add the user's approval response
            approvals.append(Message("user", [user_input_needed.to_function_approval_response(approved=True)]))

        # Run again with all the context
        result = await agent.run(approvals, session=session)
        if result.text:
            print(f"\nAgent response: {result.text}")


if __name__ == "__main__":
    asyncio.run(main())

Error Messages / Stack Traces

agent_framework.exceptions.ChatClientException: <class 'agent_framework.azure._responses_client.AzureOpenAIResponsesClient'> service failed to complete the prompt: Error code: 400 - {'error': {'message': 'No tool call found for function call output with call_id call_vkdrwm8sXNDMP2xsnZRFJx7H.', 'type': 'invalid_request_error', 'param': 'input', 'code': None}}

Package Versions

agent-framework

Python Version

3.10

Additional Context

No response