Agentscope Workstation 2.0 (#642)

Author: qbc2016

README.md

@@ -58,6 +58,7 @@ Easy for beginners, powerful for experts.
 - **Developer-friendly**: Low-code development, visual tracing & monitoring. From developing to deployment, all in one place.
 
 ## 📢 News
+- **[2025-07-01]** A new version AgentScope is under development. In this new version, AgentScope will be more powerful and flexible, with a new architecture and more features. Refer to our [Roadmap](https://github.com/modelscope/agentscope/blob/main/docs/ROADMAP.md) for more details!
 - **[2025-04-27]** A new 💻 AgentScope Studio is online now. Refer [here](https://doc.agentscope.io/build_tutorial/visual.html) for more details.
 - **[2025-03-21]** AgentScope supports hooks functions now. Refer to our [tutorial](https://doc.agentscope.io/build_tutorial/hook.html) for more details.
 - **[2025-03-19]** AgentScope supports 🔧 tools API now. Refer to our [tutorial](https://doc.agentscope.io/build_tutorial/tool.html).

README_JA.md

@@ -58,6 +58,7 @@
 - **開発者フレンドリー**：ローコード開発、視覚的なトレース＆モニタリング。開発からデプロイまで、すべてが一箇所で完結。
 
 ## 📢 ニュース
+- **[2025-07-01]** 新しいバージョンのAgentScopeが開発中です。この新しいバージョンでは、AgentScopeはより強力で柔軟になり、新しいアーキテクチャとより多くの機能を備えています。詳細については、[ロードマップ](https://github.com/modelscope/agentscope/blob/main/docs/ROADMAP.md)を参照してください！
 - **[2025-04-27]** 新しい 💻 AgentScope Studioが公開されました。詳細は[こちら](https://doc.agentscope.io/build_tutorial/visual.html)をご覧ください。
 - **[2025-03-21]** AgentScopeはフック関数をサポートしました。詳細は[チュートリアル](https://doc.agentscope.io/build_tutorial/hook.html)をご覧ください。
 - **[2025-03-19]** AgentScopeは 🔧 ツールAPIをサポートしました。詳細は[チュートリアル](https://doc.agentscope.io/build_tutorial/tool.html)をご覧ください。

README_ZH.md

@@ -58,6 +58,7 @@
 - **开发者友好**：低代码开发，可视化追踪和监控。从开发到部署，一站式解决。
 
 ## 📢 新闻
+- **[2025-07-01]** 新版本的 AgentScope 正在开发中，在这个新的版本中，AgentScope 将变的更加强大和灵活，具有更多新的功能和架构，目前[开发路线图](https://github.com/modelscope/agentscope/blob/main/docs/ROADMAP.md)已更新。
 - **[2025-04-27]** 新的 💻 AgentScope Studio 现已上线。详情请参考[链接](https://doc.agentscope.io/build_tutorial/visual.html)。
 - **[2025-03-21]** AgentScope 现已支持钩子函数。详情请参考[链接](https://doc.agentscope.io/build_tutorial/hook.html)。
 - **[2025-03-19]** AgentScope 现在支持 🔧 Tools API。详情请参考[链接](https://doc.agentscope.io/build_tutorial/tool.html)。

docs/ROADMAP.md

@@ -6,64 +6,30 @@ Offering **agent-oriented programming (AOP)** as a new programming model to orga
 
 ## Short-term Goals
 
-1. Refine and improve the documentation for easier understanding.
-2. Support tools API.
-3. Refactor the current AgentScope studio, including Dashboard and Workstation.
-4. Improve the current RAG module.
-5. Re-factor the `AgentBase` class to support more flexible serialization.
+### AgentScope V0.2.0 Roadmap
 
-## Task
+We are deeply grateful for the continuous support from the open-source community that has witnessed AgentScope's
+growth. Throughout our journey, we have maintained **developer-centric transparency** as our core principle,
+which will continue to guide our future development.
 
-1. Documentation
+As the AI agent ecosystem rapidly evolves, we recognize the need to adapt AgentScope to meet emerging trends and
+requirements. We are excited to announce the upcoming release of AgentScope v0.2.0, which marks a significant shift
+towards deployment-oriented development. This new version will provide comprehensive support for agent developers
+with enhanced deployment capabilities and practical features. Specifically, the update will include:
 
- - ✅ Re-write the tutorial.
- - 📝 Correct the typographical errors in API documents.
- - 🚧 Refine the README.md.
+- Deployment-oriented
+  - Comprehensive and native asynchronous execution
+  - Automatic session/state management
 
-2. Tools Calling
+- Developer-oriented
+  - Production-level monitoring and tracing
 
- - ✅ Add Support for Anthropic API.
+- Advanced tools management
+  - Stateful tools and toolkits management
 
- - 🚧 Support tools calling in user-assistant conversations.
-   - ✅ OpenAI API
-   - ✅ DashScope API
-   - ✅ Anthropic API
-   - 📝 Gemini APi
+- Agent-oriented evaluation
+  - Extensive and effective evaluation framework for agents
 
- - 📝 Support tools calling in multi-agent conversations.
-   - OpenAI API
-   - DashScope API
-   - Anthropic API
-   - Gemini API
-
-> 💡**Note:** The most difficult part of supporting tools calling in multi-agent conversations is most LLM APIs only support
-> `"user"` and `"assistant"` in their role fields, and has special requirements (e.g. user and assistant messages must be alternated).
-> We are working on a solution to be compatible with tools calling in
-> multi-agent conversations. If you have any ideas, please let us know. Discord | Dingtalk | GitHub issue are all welcome 🤗! Thanks in advance!
-
- - 📝 Support tools calling in streaming mode.
-   - OpenAI API
-   - DashScope API
-   - Anthropic API
-   - Gemini API
-
-3. AgentScope Studio
-
- - 🚧 Refactor the AgentScope Workstation with React.
- - 📝 Refactor the AgentScope Dashboard with React.
-   - Support websocket re-connection.
-   - Support displaying token usage.
-   - Support displaying real-time memory of agents.
-
-4. RAG
-
- - 🚧 Provide a set of query rewrite strategies for retrieval.
- - ✅ Support online search engine (Google/Bing search) based RAG.
- - 🚧 Support multi-RAG agent routing efficiently.
-
-5. AgentBase
-
- - 🚧 Decouple the data and service layer of the `MemoryBase` class to support more customized implementations.
- - 📝 Provide a new base class `Serializable` to
-   - support automatic serialization/deserialization of agents.
-   - support state registration
+Stay tuned for our detailed release notes and beta version, which will be available soon. Follow our GitHub
+repository and official channels for the latest updates. We look forward to your valuable feedback and continued
+support in shaping the future of AgentScope.

docs/news_en.md

@@ -43,7 +43,7 @@
 
 - **[2024-04-06]** We release **AgentScope** v0.0.3 now!
 
-- **[2024-04-06]** New examples [Gomoku](https://github.com/modelscope/agentscope/blob/main/examples/game_gomoku), [Conversation with ReAct Agent](https://github.com/modelscope/agentscope/blob/main/examples/conversation_with_react_agent), [Conversation with RAG Agent](https://github.com/modelscope/agentscope/blob/main/examples/conversation_with_RAG_agents) and [Distributed Parallel Optimization](https://github.com/modelscope/agentscope/blob/main/examples/distributed_parallel_optimization) are available now!
+- **[2024-04-06]** New examples [Gomoku](https://github.com/modelscope/agentscope/blob/main/examples/game_gomoku), [Conversation with ReAct Agent](https://github.com/modelscope/agentscope/blob/main/examples/agent_ReAct), [Conversation with RAG Agent](https://github.com/modelscope/agentscope/blob/main/examples/conversation_with_RAG_agents) and [Distributed Parallel Optimization](https://github.com/modelscope/agentscope/blob/main/examples/distributed_parallel_optimization) are available now!
 
 - **[2024-03-19]** We release **AgentScope** v0.0.2 now! In this new version,
 AgentScope supports ollama(A local CPU inference engine), DashScope and Google Gemini APIs.

docs/news_ja.md

@@ -34,7 +34,7 @@
 
 - **[2024-04-06]** **AgentScope** v0.0.3がリリースされました。
 
-- **[2024-04-06]** 新しい例[五目並べ](https://github.com/modelscope/agentscope/blob/main/examples/game_gomoku)、[ReActエージェントとの会話](https://github.com/modelscope/agentscope/blob/main/examples/conversation_with_react_agent)、[RAGエージェントとの会話](https://github.com/modelscope/agentscope/blob/main/examples/conversation_with_RAG_agents)、および[分散並列最適化](https://github.com/modelscope/agentscope/blob/main/examples/distributed_parallel_optimization)が利用可能になりました。
+- **[2024-04-06]** 新しい例[五目並べ](https://github.com/modelscope/agentscope/blob/main/examples/game_gomoku)、[ReActエージェントとの会話](https://github.com/modelscope/agentscope/blob/main/examples/agent_ReAct)、[RAGエージェントとの会話](https://github.com/modelscope/agentscope/blob/main/examples/conversation_with_RAG_agents)、および[分散並列最適化](https://github.com/modelscope/agentscope/blob/main/examples/distributed_parallel_optimization)が利用可能になりました。
 
 - **[2024-03-19]** **AgentScope** v0.0.2がリリースされました。この新しいバージョンでは、AgentScopeはollama（ローカルCPU推論エンジン）、DashScopeおよびGoogle Gemini APIをサポートしています。
 

docs/news_zh.md

@@ -34,7 +34,7 @@
 
 - **[2024-04-06]** 我们现在发布了**AgentScope** v0.0.3版本！
 
-- **[2024-04-06]** 新的样例“[五子棋](./examples/game_gomoku)”，“[与ReAct智能体对话](./examples/conversation_with_react_agent)”，“[与RAG智能体对话](./examples/conversation_with_RAG_agents)”，“[分布式并行优化](./examples/distributed_parallel_optimization)”上线了！
+- **[2024-04-06]** 新的样例“[五子棋](./examples/game_gomoku)”，“[与ReAct智能体对话](./examples/agent_ReAct)”，“[与RAG智能体对话](./examples/conversation_with_RAG_agents)”，“[分布式并行优化](./examples/distributed_parallel_optimization)”上线了！
 
 - **[2024-03-19]** 我们现在发布了**AgentScope** v0.0.2版本！在这个新版本中，AgentScope支持了ollama（本地CPU推理引擎），DashScope和Gemini APIs。
 

examples/agent_ReAct/README.md

@@ -0,0 +1,19 @@
+# ReAct Agent in AgentScope
+
+This example demonstrates how to use the ReAct agent in AgentScope to build a
+simple conversation.
+
+**Tip:** AgentScope provides two implementations of ReAct agent:
+- `agentscope.agents.ReActAgent`: Extract tool calls from the text response of LLMs locally.
+- `agentscope.agents.ReActAgentV2`: Use the tools API, where the tool calls extraction is done by the LLM API provider, currently supports OpenAI, DashScope, Anthropic.
+
+They share the same interface, so you can easily switch between them. The
+example in `main.py` use `ReActAgent` by default, you can change it to
+`ReActAgentV2` by modifying the import statement.
+
+
+To run this example, install agentscope and then execute the following command:
+
+```bash
+python main.py
+```

…nt/code/conversation_with_react_agent.py examples/agent_ReAct/main.pyexamples/conversation_with_react_agent/code/conversation_with_react_agent.py renamed to examples/agent_ReAct/main.py examples/conversation_with_react_agent/code/conversation_with_react_agent.py renamed to examples/agent_ReAct/main.py

@@ -1,7 +1,5 @@
 # -*- coding: utf-8 -*-
 """An example of a conversation with a ReAct agent."""
-import sys
-import io
 
 from agentscope.agents import UserAgent
 from agentscope.agents import ReActAgent
@@ -10,8 +8,7 @@
     read_text_file,
     write_text_file,
     ServiceToolkit,
-    ServiceResponse,
-    ServiceExecStatus,
+    execute_python_code,
 )
 import agentscope
 
@@ -25,41 +22,6 @@
     # ...
 }
 
-
-# Prepare a new tool function
-def execute_python_code(code: str) -> ServiceResponse:  # pylint: disable=C0301
-    """
-    Execute Python code and capture the output. Note you must `print` the output to get the result.
-    Args:
-        code (`str`):
-            The Python code to be executed.
-    """  # noqa
-
-    # Create a StringIO object to capture the output
-    old_stdout = sys.stdout
-    new_stdout = io.StringIO()
-    sys.stdout = new_stdout
-
-    try:
-        # Using `exec` to execute code
-        namespace = {}
-        exec(code, namespace)
-    except Exception as e:
-        # If an exception occurs, capture the exception information
-        output = str(e)
-        status = ServiceExecStatus.ERROR
-    else:
-        # If the execution is successful, capture the output
-        output = new_stdout.getvalue()
-        status = ServiceExecStatus.SUCCESS
-    finally:
-        # Recover the standard output
-        sys.stdout = old_stdout
-
-    # Wrap the output and status into a ServiceResponse object
-    return ServiceResponse(status, output)
-
-
 # Prepare the tools for the agent
 service_toolkit = ServiceToolkit()
 
@@ -71,7 +33,6 @@ def execute_python_code(code: str) -> ServiceResponse:  # pylint: disable=C0301
 agentscope.init(
     model_configs=YOUR_MODEL_CONFIGURATION,
     project="Conversation with ReActAgent",
-    save_api_invoke=True,
 )
 
 # Create agents

examples/conversation_nl2sql/react_nl2sql.ipynb

@@ -21,7 +21,7 @@
     "- Follow [READMD.md](https://github.com/modelscope/agentscope) to install AgentScope.\n",
     "- Install the third-party libraries used in [DAIL-SQL](https://github.com/BeachWang/DAIL-SQL/blob/main/requirements.txt).\n",
     "- Prepare a model configuration. AgentScope supports both local deployed model services (CPU or GPU) and third-party services. More details and example model configurations please refer to our [tutorial](https://doc.agentscope.io/build_tutorial/model.html).\n",
-    "- Get familiar with service functions, ServiceToolkit module and built-in ReAct agent. The corresponding example can be found in [ReAct agent example](../conversation_with_react_agent/main.ipynb).\n",
+    "- Get familiar with service functions, ServiceToolkit module and built-in ReAct agent. The corresponding example can be found in [ReAct agent example](../agent_ReAct/main.ipynb).\n",
     "\n",
     "## Note\n",
     "\n",