.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "tutorial/workflow_handoffs.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        :ref:`Go to the end <sphx_glr_download_tutorial_workflow_handoffs.py>`
        to download the full example code.

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_tutorial_workflow_handoffs.py:


.. _handoffs:

Handoffs
========================================

Handoffs 是由 OpenAI 提出的工作流模式，通过调用子智能体的方式来完成目标任务。
在 AgentScope 中通过工具调用的方式实现 handoffs 非常简单。首先，我们创建一个函数来允许协调者动态创建子智能体。

.. figure:: ../../_static/images/handoffs.png
   :width: 80%
   :align: center
   :alt: 协调者-子智能体工作流

   *Handoffs 示例*

.. GENERATED FROM PYTHON SOURCE LINES 19-95




.. rst-class:: sphx-glr-script-out

 .. code-block:: none

    Orchestrator: {
        "type": "tool_use",
        "name": "create_worker",
        "input": {
            "task_description": "在 Python 中编写并执行一个简单的 'Hello, World!' 程序。"
        },
        "id": "call_096fe6c88b904f6f9ec132"
    }
    Worker: {
        "type": "tool_use",
        "name": "execute_python_code",
        "input": {
            "code": "print('Hello, World!')",
            "timeout": 5
        },
        "id": "call_c8ebe266e0a0440598728f"
    }
    system: {
        "type": "tool_result",
        "id": "call_c8ebe266e0a0440598728f",
        "name": "execute_python_code",
        "output": [
            {
                "type": "text",
                "text": "<returncode>0</returncode><stdout>Hello, World!\n</stdout><stderr></stderr>"
            }
        ]
    }
    Worker: 这里是您请求的“Hello, World!” Python程序的执行输出：

    ```
    Hello, World!
    ```
    system: {
        "type": "tool_result",
        "id": "call_096fe6c88b904f6f9ec132",
        "name": "create_worker",
        "output": [
            {
                "type": "text",
                "text": "这里是您请求的“Hello, World!” Python程序的执行输出：\n\n```\nHello, World!\n```"
            }
        ]
    }
    Orchestrator: 子智能体已经成功地在 Python 中执行了 "Hello, World!" 程序。输出如下：

    ```
    Hello, World!
    ```






|

.. code-block:: Python


    import asyncio
    import os

    from agentscope.agent import ReActAgent
    from agentscope.formatter import DashScopeChatFormatter
    from agentscope.memory import InMemoryMemory
    from agentscope.message import Msg
    from agentscope.model import DashScopeChatModel
    from agentscope.tool import (
        ToolResponse,
        Toolkit,
        execute_python_code,
    )


    # 创建子智能体的工具函数
    async def create_worker(
        task_description: str,
    ) -> ToolResponse:
        """创建一个子智能体来完成给定的任务。子智能体配备了 Python 执行工具。

        Args:
            task_description (``str``):
                子智能体要完成的任务描述。
        """
        # 为子智能体智能体配备一些工具
        toolkit = Toolkit()
        toolkit.register_tool_function(execute_python_code)

        # 创建子智能体智能体
        worker = ReActAgent(
            name="Worker",
            sys_prompt="你是一个智能体。你的目标是完成给定的任务。",
            model=DashScopeChatModel(
                model_name="qwen-max",
                api_key=os.environ["DASHSCOPE_API_KEY"],
                stream=False,
            ),
            formatter=DashScopeChatFormatter(),
            toolkit=toolkit,
        )
        # 让子智能体完成任务
        res = await worker(Msg("user", task_description, "user"))
        return ToolResponse(
            content=res.get_content_blocks("text"),
        )


    async def run_handoffs() -> None:
        """交接工作流示例。"""
        # 初始化协调者智能体
        toolkit = Toolkit()
        toolkit.register_tool_function(create_worker)

        orchestrator = ReActAgent(
            name="Orchestrator",
            sys_prompt="你是一个协调者智能体。你的目标是通过将任务分解为更小的任务并创建子智能体来完成它们，从而完成给定的任务。",
            model=DashScopeChatModel(
                model_name="qwen-max",
                api_key=os.environ["DASHSCOPE_API_KEY"],
                stream=False,
            ),
            memory=InMemoryMemory(),
            formatter=DashScopeChatFormatter(),
            toolkit=toolkit,
        )

        # 任务描述
        task_description = "在 Python 中执行 hello world"

        # 创建子智能体来完成任务
        await orchestrator(Msg("user", task_description, "user"))


    asyncio.run(run_handoffs())


.. rst-class:: sphx-glr-timing

   **Total running time of the script:** (0 minutes 12.880 seconds)


.. _sphx_glr_download_tutorial_workflow_handoffs.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: workflow_handoffs.ipynb <workflow_handoffs.ipynb>`

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: workflow_handoffs.py <workflow_handoffs.py>`

    .. container:: sphx-glr-download sphx-glr-download-zip

      :download:`Download zipped: workflow_handoffs.zip <workflow_handoffs.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_