ag-ui-protocol/ag-ui

Project overview

It addresses the lack of a standard connection layer between AI agents and frontend applications, and ships a reference HTTP implementation plus partnership integrations with LangChain/LangGraph and CrewAI, giving developers a concrete starting path rather than only a specification.

Project type

AI Agent

Use cases

Automation

Deployment

Refer to project documentation

License

MIT

Best for

• Developers integrating AI agents with user-facing frontend applications who need a standardized, event-based connection layer.

Key capabilities

• An event-based protocol with ~16 standard event types that agent backends emit, and simple compatible inputs they accept.
• Ships a reference HTTP implementation and a default connector to help teams get started fast.

Getting started

• Run `npx create-ag-ui-app my-agent-app`, then follow the AG-UI quickstart docs. Use the reference HTTP implementation and default connector to get started. Setup difficulty is documented as easy, but coding is required.

Evidence and sources

• README: AG-UI is an open, lightweight, event-based protocol that standardizes how AI agents connect to user-facing applications.
• README: it enables seamless integration between AI agents, real time user context, and user interfaces.
• README: agent backends **emit events _compatible_ with one of AG-UI's ~16 standard event types**
• README: It also ships with a **reference HTTP implementation** and **default connector** to help teams get started fast.
• README: npx create-ag-ui-app my-agent-app

Source updated at: 2026-10-19T18:36:08Z

Open original GitHub project