ag-ui README

The project's README from github.com/ag-ui-protocol/ag-ui. As one markdown file: README.md.

ag-ui Logo AG-UI: The Agent-User Interaction Protocol

AG-UI is an open, lightweight, event-based protocol that standardizes how AI agents connect to user-facing applications. Built for simplicity and flexibility, it enables seamless integration between AI agents, real time user context, and user interfaces.



Version MIT Discord

Join our Discord →     Read the Docs →     Go to the AG-UI Dojo →     Follow us →

1600x680

🚀 Getting Started

Create a new AG-UI application in seconds:

npx create-ag-ui-app my-agent-app

Useful Links:

What is AG-UI?

AG-UI is an open, lightweight, event-based protocol for agent-human interaction, designed for simplicity & flexibility:

AG-UI includes a flexible middleware layer that ensures compatibility across diverse environments:

It also ships with a reference HTTP implementation and default connector to help teams get started fast.

Learn more about the specs →

Why AG-UI?

AG-UI was developed based on real-world requirements and practical experience building in-app agent interactions.

Where does AGUI fit in the agentic protocol stack?

AG-UI is complementary to the other 2 top agentic protocols

The Agent Protocol Stack

🚀 Features

🛠 Supported Integrations

AG-UI was born from CopilotKit's initial partnership with LangChain and CrewAI - and brings the incredibly popular agent-user-interactivity infrastructure to the wider agentic ecosystem.

1st party = the platforms that have AG‑UI built in and provide documentation for guidance.

Frameworks

Framework Status AG-UI Resources
Built-in Agent ✅ Supported ➡️ Docs

🤝 Partnerships

Framework Status AG-UI Resources
LangChain ✅ Supported ➡️ Docs 🎮 Demos
CrewAI ✅ Supported ➡️ Docs 🎮 Demos

🧩 1st Party

Framework Status AG-UI Resources
Microsoft Agent Framework ✅ Supported ➡️ Docs 🎮 Demos
Google ADK ✅ Supported ➡️ Docs 🎮 Demos
AWS Strands Agents ✅ Supported ➡️ Docs 🎮 Demos
Mastra ✅ Supported ➡️ Docs 🎮 Demos
Pydantic AI ✅ Supported ➡️ Docs 🎮 Demos
Agno ✅ Supported ➡️ Docs 🎮 Demos
LlamaIndex ✅ Supported ➡️ Docs 🎮 Demos
AG2 ✅ Supported ➡️ Docs 🎮 Demos
AWS Bedrock Agents 🛠️ In Progress –

🌐 Community

Framework Status AG-UI Resources
Claude Agent SDK ✅ Supported 🎮 Demos
Claude Managed Agents SDK ✅ Supported 🎮 Demos
Langroid ✅ Supported 🎮 Demos
OpenAI Agent SDK 🛠️ In Progress –
Cloudflare Agents 🛠️ In Progress –

Agent Interaction Protocols

Protocols Status AG-UI Resources Integrations
A2A ✅ Supported ➡️ Docs Partnership

Infrastructure / Deployment

Platform Status AG-UI Resources Integrations
Amazon Bedrock AgentCore ✅ Supported ➡️ Docs 1st Party

Specification (standard)

Framework Status AG-UI Resources
Oracle Agent Spec ✅ Supported ➡️ Docs 🎮 Demos

Generative UI

Framework Status AG-UI Resources
MCP Apps ✅ Supported ➡️ Docs 🎮 Demos

SDKs

SDK Status AG-UI Resources Integrations
Kotlin ✅ Supported ➡️ Getting Started Community
Golang ✅ Supported ➡️ Getting Started Community
Dart ✅ Supported ➡️ Getting Started Community
Java ✅ Supported ➡️ Getting Started Community
Rust ✅ Supported ➡️ Getting Started Community
Ruby ✅ Supported ➡️ Getting Started Community
C++ ✅ Supported ➡️ GitHub Source Community
.NET ✅ Supported ➡️ Getting Started Community
Nim 🛠️ In Progress ➡️ PR Community
Flowise 🛠️ In Progress ➡️ GitHub Source Community
Langflow 🛠️ In Progress ➡️ GitHub Source Community

Clients

Client Status AG-UI Resources Integrations
CopilotKit ✅ Supported ➡️ Getting Started 1st Party
Terminal + Agent ✅ Supported ➡️ Getting Started Community
Chat platforms (Slack, Microsoft Teams) ✅ Supported ➡️ Channels SDK 🎮 OpenTag example 1st Party
React Native ✅ Supported ➡️ Example 1st Party

View all supported integrations →

Examples

Hello World App

Video:

https://github.com/user-attachments/assets/18c03330-1ebc-4863-b2b8-cc6c3a4c7bae

https://agui-demo.vercel.app/

The AG-UI Dojo (Building-Blocks Viewer)

The AG-UI Dojo demonstrates AG-UI's core building blocks through simple, focused examples—each just 50-200 lines of code.

View the source code for the Dojo and all framework integrations here.

https://github.com/user-attachments/assets/c298eea8-3f39-4a94-b968-7712429b0c49

🙋🏽‍♂️ Contributing to AG-UI

Check out the Contributing guide

Roadmap

Check out the AG-UI Roadmap to see what's being built and where you can jump in.

📄 License

AG-UI is open source software licensed as MIT.

ag-ui (ag-ui-protocol/ag-ui)

Searching Database…

16.7K ·GitHub
这是什么?AG-UI 是一个开放、轻量、基于事件的协议,用来规范 AI 智能体如何连接到面向用户的应用——它相当于智能体后端和展示其工作过程的界面之间的一份通用事件约定。
它是一份要阅读的规范,还是可以安装的代码?它是一份协议规范:智能体后端按照大约 16 种标准事件类型发出事件,中间件层让这些事件可以通过任何传输方式传递——SSE、WebSocket、webhook 都行——仓库里还附带了 SDK 和一个参考 HTTP 实现。
我可以接入哪些智能体框架?官方和合作伙伴支持的有 LangChain、CrewAI、Microsoft Agent Framework、Google ADK、AWS Strands Agents、Mastra、Pydantic AI、Agno、LlamaIndex 和 AG2;社区集成包括 Claude Agent SDK 和 Langroid。
还有哪些集成尚未完成?框架方面,AWS Bedrock Agents、OpenAI Agent SDK 和 Cloudflare Agents 还没完成;另外 Nim SDK 以及 Flowise 和 Langflow 的集成也尚未完成。
智能体的界面实际能运行在哪里?只要能做客户端的地方都可以:CopilotKit 网页应用、终端加智能体的组合、Slack 和 Microsoft Teams 这类聊天平台,以及 React Native 移动应用,都是文档里写明的客户端。
基于 AG-UI 构建的应用到底能做什么?支持流式输出的实时智能体对话、双向状态同步、基于结构化消息的生成式界面、实时上下文增强、前端工具集成,以及人在回路的协作。
AG-UI 和 MCP、A2A 是什么关系?它们是智能体技术栈中互补的几个层次:MCP 给智能体提供工具,A2A 让智能体之间互相通信,而 AG-UI 把智能体接入面向用户的应用。
除了 TypeScript,还有其他语言的 SDK 吗?有——社区 SDK 覆盖 Kotlin、Golang、Dart、Java、Rust、Ruby、C++ 和 .NET,Nim 的还在开发中,所以你的智能体后端可以继续用它原本的技术栈。
AG-UI Dojo 是什么?它是这个项目的功能演示集:用简单、专注的示例展示 AG-UI 的各项核心功能,每个示例只有 50 到 200 行代码,全部源码都在仓库里——是单独查看每个功能的最快方式。
What are similar projects?CopilotKit, tambo, openui, LibreChat, boardui, ai, taskuary, deer-flow
Size
76.5 MB
Stars
16.7K
Age
1y
Commits · 90 Days