base on Start building LLM-empowered multi-agent applications in an easier way. [**中文主页**](https://github.com/modelscope/agentscope/blob/main/README_ZH.md) | [**日本語のホームページ**](https://github.com/modelscope/agentscope/blob/main/README_JA.md) | [**Tutorial**](https://doc.agentscope.io/) | [**Roadmap**](https://github.com/modelscope/agentscope/blob/main/docs/ROADMAP.md) | [**FAQ**](https://doc.agentscope.io/tutorial/faq.html) <h2 align="center">AgentScope: Agent-Oriented Programming for Building LLM Applications</h2> <p align="center"> <a href="https://arxiv.org/abs/2402.14034"> <img src="https://img.shields.io/badge/cs.MA-2402.14034-B31C1C?logo=arxiv&logoColor=B31C1C" alt="arxiv" /> </a> <a href="https://pypi.org/project/agentscope/"> <img src="https://img.shields.io/badge/python-3.9+-blue?logo=python" alt="pypi" /> </a> <a href="https://pypi.org/project/agentscope/"> <img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fpypi.org%2Fpypi%2Fagentscope%2Fjson&query=%24.info.version&prefix=v&logo=pypi&label=version" alt="pypi" /> </a> <a href="https://doc.agentscope.io/"> <img src="https://img.shields.io/badge/Docs-English%7C%E4%B8%AD%E6%96%87-blue?logo=markdown" alt="docs" /> </a> <a href="https://agentscope.io/"> <img src="https://img.shields.io/badge/Drag_and_drop_UI-WorkStation-blue?logo=html5&logoColor=green&color=dark-green" alt="workstation" /> </a> <a href="./LICENSE"> <img src="https://img.shields.io/badge/license-Apache--2.0-black" alt="license" /> </a> </p> <p align="center"> <img src="https://trendshift.io/api/badge/repositories/10079" alt="modelscope%2Fagentscope | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/> </p> ## ✨ Why AgentScope? Easy for beginners, powerful for experts. - **Transparent to Developers**: Transparent is our **FIRST principle**. Prompt engineering, API invocation, agent building, workflow orchestration, all are visible and controllable for developers. No deep encapsulation or implicit magic. - **Model Agnostic**: Programming once, run with all models. More than **17+** LLM API providers are supported. - **LEGO-style Agent Building**: All components are **modular** and **independent**. Use them or not, your choice. - **Multi-Agent Oriented**: Designed for **multi-agent**, **explicit** message passing and workflow orchestration, NO deep encapsulation. - **Native Distribution/Parallelization**: Centralized programming for distributed application, and **automatic parallelization**. - **Highly Customizable**: Tools, prompt, agent, workflow, third-party libs & visualization, customization is encouraged everywhere. - **Developer-friendly**: Low-code development, visual tracing & monitoring. From developing to deployment, all in one place. ## 📢 News - **[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). - **[2025-03-20]** Agentscope now supports [MCP Server](https://github.com/modelcontextprotocol/servers)! You can learn how to use it by following this [tutorial](https://doc.agentscope.io/build_tutorial/MCP.html). - **[2025-03-05]** Our [🎓 AgentScope Copilot](applications/multisource_rag_app/README.md), a multi-source RAG application is open-source now! - **[2025-02-24]** [🇨🇳 Chinese version tutorial](https://doc.agentscope.io/zh_CN) is online now! - **[2025-02-13]** We have released the [📁 technical report](https://doc.agentscope.io/tutorial/swe.html) of our solution in [SWE-Bench(Verified)](https://www.swebench.com/)! - **[2025-02-07]** 🎉🎉 AgentScope has achieved a **63.4% resolve rate** in [SWE-Bench(Verified)](https://www.swebench.com/). - **[2025-01-04]** AgentScope supports Anthropic API now. 👉👉 [**Older News**](https://github.com/modelscope/agentscope/blob/main/docs/news_en.md) ## 💬 Contact Welcome to join our community on | [Discord](https://discord.gg/eYMpfnkG8h) | DingTalk | |----------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------| | <img src="https://gw.alicdn.com/imgextra/i1/O1CN01hhD1mu1Dd3BWVUvxN_!!6000000000238-2-tps-400-400.png" width="100" height="100"> | <img src="https://img.alicdn.com/imgextra/i1/O1CN01LxzZha1thpIN2cc2E_!!6000000005934-2-tps-497-477.png" width="100" height="100"> | <!-- START doctoc generated TOC please keep comment here to allow auto update --> <!-- DON'T EDIT THIS SECTION, INSTEAD RE-RUN doctoc TO UPDATE --> ## 📑 Table of Contents - [🚀 Quickstart](#-quickstart) - [💻 Installation](#-installation) - [🛠️ From source](#-from-source) - [📦 From PyPi](#-from-pypi) - [📝 Example](#-example) - [👋 Hello AgentScope](#-hello-agentscope) - [🧑‍🤝‍🧑 Multi-Agent Conversation](#-multi-agent-conversation) - [💡 Reasoning with Tools & MCP](#-reasoning-with-tools--mcp) - [🔠 Structured Output](#-structured-output) - [✏️ Workflow Orchestration](#-workflow-orchestration) - [⚡️ Distribution and Parallelization](#%EF%B8%8F-distribution-and-parallelization) - [👀 Tracing & Monitoring](#-tracing--monitoring) - [⚖️ License](#-license) - [📚 Publications](#-publications) - [✨ Contributors](#-contributors) <!-- END doctoc generated TOC please keep comment here to allow auto update --> ## 🚀 Quickstart ### 💻 Installation > AgentScope requires **Python 3.9** or higher. #### 🛠️ From source ```bash # Pull the source code from GitHub git clone https://github.com/modelscope/agentscope.git # Install the package in editable mode cd agentscope pip install -e . ``` #### 📦 From PyPi ```bash pip install agentscope ``` ## 📝 Example ### 👋 Hello AgentScope   Creating a basic conversation **explicitly** between **a user** and **an assistant** with AgentScope: ```python from agentscope.agents import DialogAgent, UserAgent import agentscope # Load model configs agentscope.init( model_configs=[ { "config_name": "my_config", "model_type": "dashscope_chat", "model_name": "qwen-max", } ] ) # Create a dialog agent and a user agent dialog_agent = DialogAgent( name="Friday", model_config_name="my_config", sys_prompt="You're a helpful assistant named Friday" ) user_agent = UserAgent(name="user") # Build the workflow/conversation explicitly x = None while x is None or x.content != "exit": x = dialog_agent(x) x = user_agent(x) ``` ### 🧑‍🤝‍🧑 Multi-Agent Conversation AgentScope is designed for **multi-agent** applications, offering flexible control over information flow and communication between agents.   ```python from agentscope.agents import DialogAgent from agentscope.message import Msg from agentscope.pipelines import sequential_pipeline from agentscope import msghub import agentscope # Load model configs agentscope.init( model_configs=[ { "config_name": "my_config", "model_type": "dashscope_chat", "model_name": "qwen-max", } ] ) # Create three agents friday = DialogAgent( name="Friday", model_config_name="my_config", sys_prompt="You're a helpful assistant named Friday" ) saturday = DialogAgent( name="Saturday", model_config_name="my_config", sys_prompt="You're a helpful assistant named Saturday" ) sunday = DialogAgent( name="Sunday", model_config_name="my_config", sys_prompt="You're a helpful assistant named Sunday" ) # Create a chatroom by msghub, where agents' messages are broadcast to all participants with msghub( participants=[friday, saturday, sunday], announcement=Msg("user", "Counting from 1 and report one number each time without other things", "user"), # A greeting message ) as hub: # Speak in sequence sequential_pipeline([friday, saturday, sunday], x=None) ``` ### 💡 Reasoning with Tools & MCP  Creating a reasoning agent with built-in tools and **MCP servers**! ```python from agentscope.agents import ReActAgentV2, UserAgent from agentscope.service import ServiceToolkit, execute_python_code import agentscope agentscope.init( model_configs={ "config_name": "my_config", "model_type": "dashscope_chat", "model_name": "qwen-max", } ) # Add tools toolkit = ServiceToolkit() toolkit.add(execute_python_code) # Connect to Gaode MCP server toolkit.add_mcp_servers( { "mcpServers": { "amap-amap-sse": { "url": "https://mcp.amap.com/sse?key={YOUR_GAODE_API_KEY}" } } } ) # Create a reasoning-acting agent agent = ReActAgentV2( name="Friday", model_config_name="my_config", service_toolkit=toolkit, sys_prompt="You're a helpful assistant named Friday." ) user_agent = UserAgent(name="user") # Build the workflow/conversation explicitly x = None while x is None or x.content != "exit": x = agent(x) x = user_agent(x) ``` ### 🔠 Structured Output  Specifying structured output with a Pydantic base model. ```python from agentscope.agents import ReActAgentV2 from agentscope.service import ServiceToolkit from agentscope.message import Msg from pydantic import BaseModel, Field from typing import Literal import agentscope agentscope.init( model_configs={ "config_name": "my_config", "model_type": "dashscope_chat", "model_name": "qwen-max", } ) # Create a reasoning-acting agent agent = ReActAgentV2( name="Friday", model_config_name="my_config", service_toolkit=ServiceToolkit(), max_iters=20 ) class CvModel(BaseModel): name: str = Field(max_length=50, description="The name") description: str = Field(max_length=200, description="The brief description") aget: int = Field(gt=0, le=120, description="The age of the person") class ChoiceModel(BaseModel): choice: Literal["apple", "banana"] # Specify structured output using `structured_model` res_msg = agent( Msg("user", "Introduce Einstein", "user"), structured_model=CvModel ) print(res_msg.metadata) # Switch to different structured model res_msg = agent( Msg("user", "Choice a fruit", "user"), structured_model=ChoiceModel ) print(res_msg.metadata) ``` ### ✏️ Workflow Orchestration  [Routing](https://www.anthropic.com/engineering/building-effective-agents), [parallelization](https://www.anthropic.com/engineering/building-effective-agents), [orchestrator-workers](https://www.anthropic.com/engineering/building-effective-agents), or [evaluator-optimizer](https://www.anthropic.com/engineering/building-effective-agents). Build your own workflow with AgentScope easily! Taking routing as an example: ```python from agentscope.agents import ReActAgentV2 from agentscope.service import ServiceToolkit from agentscope.message import Msg from pydantic import BaseModel, Field from typing import Literal, Union import agentscope agentscope.init( model_configs={ "config_name": "my_config", "model_type": "dashscope_chat", "model_name": "qwen-max", } ) # Workflow: Routing routing_agent = ReActAgentV2( name="Routing", model_config_name="my_config", sys_prompt="You're a routing agent. Your target is to route the user query to the right follow-up task", service_toolkit=ServiceToolkit() ) # Use structured output to specify the routing task class RoutingChoice(BaseModel): your_choice: Literal[ 'Content Generation', 'Programming', 'Information Retrieval', None ] = Field(description="Choice the right follow-up task, and choice `None` if the task is too simple or no suitable task") task_description: Union[str, None] = Field(description="The task description", default=None) res_msg = routing_agent( Msg("user", "Help me to write a poem", "user"), structured_model=RoutingChoice ) # Execute the follow-up task if res_msg.metadata["your_choice"] == "Content Generation": ... elif res_msg.metadata["your_choice"] == "Programming": ... elif res_msg.metadata["your_choice"] == "Information Retrieval": ... else: ... ``` ### ⚡️ Distribution and Parallelization    Using `to_dist` function to run the agent in distributed mode! ```python from agentscope.agents import DialogAgent from agentscope.message import Msg import agentscope # Load model configs agentscope.init( model_configs=[ { "config_name": "my_config", "model_type": "dashscope_chat", "model_name": "qwen-max", } ] ) # Using `to_dist()` to run the agent in distributed mode agent1 = DialogAgent( name="Saturday", model_config_name="my_config" ).to_dist() agent2 = DialogAgent( name="Sunday", model_config_name="my_config" ).to_dist() # The two agent will run in parallel agent1(Msg("user", "Execute task1 ...", "user")) agent2(Msg("user", "Execute task2 ...", "user")) ``` ### 👀 Tracing & Monitoring   AgentScope provides a local visualization and monitoring tool, **AgentScope Studio**. ```bash # Install AgentScope Studio npm install -g @agentscope/studio # Run AgentScope Studio as_studio ``` ```python import agentscope # Connect application to AgentScope Studio agentscope.init( model_configs = { "config_name": "my_config", "model_type": "dashscope_chat", "model_name": "qwen_max", }, studio_url="http://localhost:3000", # The URL of AgentScope Studio ) # ... ``` <div align="center"> <img src="https://img.alicdn.com/imgextra/i4/O1CN01eCEYvA1ueuOkien7T_!!6000000006063-1-tps-960-600.gif" alt="AgentScope Studio" width="100%" /> <div align="center">AgentScope Studio, a local visualization tool</div> </div> ## ⚖️ License AgentScope is released under Apache License 2.0. ## 📚 Publications If you find our work helpful for your research or application, please cite our papers. [AgentScope: A Flexible yet Robust Multi-Agent Platform](https://arxiv.org/abs/2402.14034) ``` @article{agentscope, author = {Dawei Gao and Zitao Li and Xuchen Pan and Weirui Kuang and Zhijian Ma and Bingchen Qian and Fei Wei and Wenhao Zhang and Yuexiang Xie and Daoyuan Chen and Liuyi Yao and Hongyi Peng and Ze Yu Zhang and Lin Zhu and Chen Cheng and Hongzhu Shi and Yaliang Li and Bolin Ding and Jingren Zhou} title = {AgentScope: A Flexible yet Robust Multi-Agent Platform}, journal = {CoRR}, volume = {abs/2402.14034}, year = {2024}, } ``` ## ✨ Contributors All thanks to our contributors: <a href="https://github.com/modelscope/agentscope/graphs/contributors"> <img src="https://contrib.rocks/image?repo=modelscope/agentscope&max=999&columns=12&anon=1" /> </a>", Assign "at most 3 tags" to the expected json: {"id":"10079","tags":[]} "only from the tags list I provide: [{"id":77,"name":"3d"},{"id":89,"name":"agent"},{"id":17,"name":"ai"},{"id":54,"name":"algorithm"},{"id":24,"name":"api"},{"id":44,"name":"authentication"},{"id":3,"name":"aws"},{"id":27,"name":"backend"},{"id":60,"name":"benchmark"},{"id":72,"name":"best-practices"},{"id":39,"name":"bitcoin"},{"id":37,"name":"blockchain"},{"id":1,"name":"blog"},{"id":45,"name":"bundler"},{"id":58,"name":"cache"},{"id":21,"name":"chat"},{"id":49,"name":"cicd"},{"id":4,"name":"cli"},{"id":64,"name":"cloud-native"},{"id":48,"name":"cms"},{"id":61,"name":"compiler"},{"id":68,"name":"containerization"},{"id":92,"name":"crm"},{"id":34,"name":"data"},{"id":47,"name":"database"},{"id":8,"name":"declarative-gui "},{"id":9,"name":"deploy-tool"},{"id":53,"name":"desktop-app"},{"id":6,"name":"dev-exp-lib"},{"id":59,"name":"dev-tool"},{"id":13,"name":"ecommerce"},{"id":26,"name":"editor"},{"id":66,"name":"emulator"},{"id":62,"name":"filesystem"},{"id":80,"name":"finance"},{"id":15,"name":"firmware"},{"id":73,"name":"for-fun"},{"id":2,"name":"framework"},{"id":11,"name":"frontend"},{"id":22,"name":"game"},{"id":81,"name":"game-engine "},{"id":23,"name":"graphql"},{"id":84,"name":"gui"},{"id":91,"name":"http"},{"id":5,"name":"http-client"},{"id":51,"name":"iac"},{"id":30,"name":"ide"},{"id":78,"name":"iot"},{"id":40,"name":"json"},{"id":83,"name":"julian"},{"id":38,"name":"k8s"},{"id":31,"name":"language"},{"id":10,"name":"learning-resource"},{"id":33,"name":"lib"},{"id":41,"name":"linter"},{"id":28,"name":"lms"},{"id":16,"name":"logging"},{"id":76,"name":"low-code"},{"id":90,"name":"message-queue"},{"id":42,"name":"mobile-app"},{"id":18,"name":"monitoring"},{"id":36,"name":"networking"},{"id":7,"name":"node-version"},{"id":55,"name":"nosql"},{"id":57,"name":"observability"},{"id":46,"name":"orm"},{"id":52,"name":"os"},{"id":14,"name":"parser"},{"id":74,"name":"react"},{"id":82,"name":"real-time"},{"id":56,"name":"robot"},{"id":65,"name":"runtime"},{"id":32,"name":"sdk"},{"id":71,"name":"search"},{"id":63,"name":"secrets"},{"id":25,"name":"security"},{"id":85,"name":"server"},{"id":86,"name":"serverless"},{"id":70,"name":"storage"},{"id":75,"name":"system-design"},{"id":79,"name":"terminal"},{"id":29,"name":"testing"},{"id":12,"name":"ui"},{"id":50,"name":"ux"},{"id":88,"name":"video"},{"id":20,"name":"web-app"},{"id":35,"name":"web-server"},{"id":43,"name":"webassembly"},{"id":69,"name":"workflow"},{"id":87,"name":"yaml"}]" returns me the "expected json"