:orphan:




.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="About AgentScope ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ What is AgentScope?     AgentScope is a multi-agent framework, aiming to provide a simple yet efficient way to build LLM-empowered agent applications.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_faq_thumb.png
    :alt:

  :doc:`/tutorial/faq`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">FAQ</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope implements OpenTelemetry-based tracing to monitor and debug the execution of agent applications, which features">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_tracing_thumb.png
    :alt:

  :doc:`/tutorial/task_tracing`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Tracing</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This chapter establishes key concepts from an engineering perspective to introduce AgentScope&#x27;s design.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_quickstart_key_concept_thumb.png
    :alt:

  :doc:`/tutorial/quickstart_key_concept`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Key Concepts</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="In AgentScope, the memory is used to store the context of the agent, and retrieve it when needed. Specifically, AgentScope provides a memory base class MemoryBase and an in-memory implementation InMemoryMemory under agentscope.memory that can be used directly.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_memory_thumb.png
    :alt:

  :doc:`/tutorial/task_memory`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Memory</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope Studio is a local-deployed web application that">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_studio_thumb.png
    :alt:

  :doc:`/tutorial/task_studio`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">AgentScope Studio</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope requires Python 3.10 or higher. You can install from source or pypi.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_quickstart_installation_thumb.png
    :alt:

  :doc:`/tutorial/quickstart_installation`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Installation</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope provides a token counter module under agentscope.token to calculate the number of tokens in the given messages, allowing developers to estimate the number of tokens in a prompt before sending it to an API.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_token_thumb.png
    :alt:

  :doc:`/tutorial/task_token`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Token</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="A simple example is shown below, where two agents are created and executed concurrently.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_workflow_concurrent_agents_thumb.png
    :alt:

  :doc:`/tutorial/workflow_concurrent_agents`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Concurrent Agents</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Agent skill is an approach proposed by Anthropic to improve agent capabilities on specific tasks.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_agent_skill_thumb.png
    :alt:

  :doc:`/tutorial/task_agent_skill`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Agent Skill</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Hooks are extension points in AgentScope that allow developers to customize agent behaviors at specific execution points, providing a flexible way to modify or extend the agent&#x27;s functionality without changing its core implementation.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_hook_thumb.png
    :alt:

  :doc:`/tutorial/task_hook`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Agent Hooks</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="   Handoffs example">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_workflow_handoffs_thumb.png
    :alt:

  :doc:`/tutorial/workflow_handoffs`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Handoffs</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="For multi-agent orchestration, AgentScope provides the agentscope.pipeline module as syntax sugar for chaining agents together, including">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_pipeline_thumb.png
    :alt:

  :doc:`/tutorial/task_pipeline`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Pipeline</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="In AgentScope, we provide a basic class for long-term memory (``LongTermMemoryBase``) and an implementation based on the mem0 library (``Mem0LongTermMemory``). Together with the design of ReActAgent class in agent section, we provide two long-term memory modes:">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_long_term_memory_thumb.png
    :alt:

  :doc:`/tutorial/task_long_term_memory`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Long-Term Memory</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Message is the core concept in AgentScope, used to support multimodal data, tools API, information storage/exchange and prompt construction.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_quickstart_message_thumb.png
    :alt:

  :doc:`/tutorial/quickstart_message`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Create Message</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The tutorial covers the following features of AgentScope in support of the MCP (Model Context Protocol):">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_mcp_thumb.png
    :alt:

  :doc:`/tutorial/task_mcp`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">MCP</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="In AgentScope, the embedding module provides a unified interface for vector representation generation, which features:">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_embedding_thumb.png
    :alt:

  :doc:`/tutorial/task_embedding`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Embedding</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Conversation is a design pattern that agents exchange and share information between each other, most commonly in game playing, chatbot, and multi-agent discussion scenarios.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_workflow_conversation_thumb.png
    :alt:

  :doc:`/tutorial/workflow_conversation`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Conversation</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope provides a unified interface for Text-to-Speech (TTS) models across multiple API providers. This tutorial demonstrates how to use TTS models in AgentScope.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_tts_thumb.png
    :alt:

  :doc:`/tutorial/task_tts`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">TTS</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="In this tutorial, we introduce the model APIs integrated in AgentScope, how to use them and how to integrate new model APIs. The supported model APIs and providers include:">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_model_thumb.png
    :alt:

  :doc:`/tutorial/task_model`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Model</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="A2A (Agent-to-Agent) is an open standard protocol for enabling interoperable communication between different AI agents.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_a2a_thumb.png
    :alt:

  :doc:`/tutorial/task_a2a`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">A2A Agent</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope provides out-of-the-box ReAct agent ReActAgent under agentscope.agent that can be used directly.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_quickstart_agent_thumb.png
    :alt:

  :doc:`/tutorial/quickstart_agent`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Create ReAct Agent</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="In AgentScope, the &quot;state&quot; refers to the agent status in the running application, including its current system prompt, memory, context, equipped tools, and other information that change over time.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_state_thumb.png
    :alt:

  :doc:`/tutorial/task_state`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">State/Session Management</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="In this tutorial, we first focus on introducing the ReAct agent in AgentScope, then we briefly introduce how to customize your own agent from scratch.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_agent_thumb.png
    :alt:

  :doc:`/tutorial/task_agent`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Agent</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The formatter module in AgentScope is responsible for">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_prompt_thumb.png
    :alt:

  :doc:`/tutorial/task_prompt`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Prompt Formatter</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="- Routing by structured output - Routing by tool calls">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_workflow_routing_thumb.png
    :alt:

  :doc:`/tutorial/workflow_routing`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Routing</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Debate workflow simulates a multi-turn discussion between different agents, mostly several solvers and an aggregator. Typically, the solvers generate and exchange their answers, while the aggregator collects and summarizes the answers.">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_workflow_multiagent_debate_thumb.png
    :alt:

  :doc:`/tutorial/workflow_multiagent_debate`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Multi-Agent Debate</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The Plan Module enables agents to formally break down complex tasks into manageable sub-tasks and execute them systematically. Key features include:">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_plan_thumb.png
    :alt:

  :doc:`/tutorial/task_plan`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Plan</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope provides a built-in evaluation framework for assessing agent performance across different tasks and benchmarks, featuring:">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_eval_thumb.png
    :alt:

  :doc:`/tutorial/task_eval`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Evaluation</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="AgentScope provides built-in support for Retrieval-Augmented Generation (RAG) tasks. This tutorial demonstrates">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_rag_thumb.png
    :alt:

  :doc:`/tutorial/task_rag`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">RAG</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="To ensure accurate and reliable tool parsing, AgentScope fully embraces the use of tools API with the following features:">

.. only:: html

  .. image:: /tutorial/images/thumb/sphx_glr_task_tool_thumb.png
    :alt:

  :doc:`/tutorial/task_tool`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Tool</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /tutorial/faq
   /tutorial/task_tracing
   /tutorial/quickstart_key_concept
   /tutorial/task_memory
   /tutorial/task_studio
   /tutorial/quickstart_installation
   /tutorial/task_token
   /tutorial/workflow_concurrent_agents
   /tutorial/task_agent_skill
   /tutorial/task_hook
   /tutorial/workflow_handoffs
   /tutorial/task_pipeline
   /tutorial/task_long_term_memory
   /tutorial/quickstart_message
   /tutorial/task_mcp
   /tutorial/task_embedding
   /tutorial/workflow_conversation
   /tutorial/task_tts
   /tutorial/task_model
   /tutorial/task_a2a
   /tutorial/quickstart_agent
   /tutorial/task_state
   /tutorial/task_agent
   /tutorial/task_prompt
   /tutorial/workflow_routing
   /tutorial/workflow_multiagent_debate
   /tutorial/task_plan
   /tutorial/task_eval
   /tutorial/task_rag
   /tutorial/task_tool



.. only:: html

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

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