a2ui

Generates declarative UIs for AI agents using the A2UI protocol.

3|Updated Jan 28, 2026
One-click install
npx skills add https://github.com/ldmrepo/michael --skill a2ui
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: a2ui
Source: https://github.com/ldmrepo/michael/tree/main/.claude/skills/a2ui
Command: npx skills add https://github.com/ldmrepo/michael --skill a2ui

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI agents often struggle to render rich, interactive UIs that adapt to user input and platform. A2UI provides a protocol and tooling to declaratively define surfaces, components, and data models that stream to clients in real time.

Core Features & Use Cases

  • Declarative UI streaming: emit surface and component definitions as JSONL to render across platforms.
  • Data binding: use BoundValue-like bindings to connect UI to a dynamic data model.
  • A2A integration: plug into AI agent workflows via a standard extension catalog.

Quick Start

Use the A2UI builder to define a surface with a title, a Card containing text and a button, then emit surfaceUpdate, dataModelUpdate, and beginRendering messages.

Frequently Asked Questions about a2ui

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate declarative UIs that stream to multiple platforms?

A2UI generates declarative UI definitions in JSONL format that stream across web, mobile, and native platforms. Define surfaces and components once, then emit them as JSON messages that each renderer interprets natively, eliminating platform-specific UI code.

Can I bind dynamic data to UI components in a streaming interface?

Yes. A2UI uses BoundValue resolution to connect UI elements to a dynamic data model. As data changes, emit dataModelUpdate messages to keep components synchronized without re-rendering the entire surface.

How do I integrate A2UI into an AI agent workflow?

A2UI integrates into agent workflows through a standard extension catalog and A2A protocol integration. Define your surfaces declaratively, emit them as JSONL, and plug the catalog into your agent's action set to enable real-time interactive interfaces.

What format does A2UI use to stream UI definitions to clients?

A2UI streams UI definitions as JSONL, a safe, framework-agnostic format that emits surface updates, component definitions, and data model changes as newline-delimited JSON messages clients can parse and render incrementally.

Does A2UI handle user actions and form input from rendered surfaces?

Yes. A2UI manages user actions through its declarative protocol. Components emit user interactions back to the agent, enabling bidirectional communication between the rendered UI and the backend without manual event handling.

Can I use A2UI without adopting a specific UI framework?

Yes. A2UI is framework-agnostic. It defines surfaces and components declaratively in a protocol-based format, leaving rendering implementation to platform-specific clients—web, mobile, or native—without framework lock-in.