a2ui

When this skill is activated, always start your first response with the 🧢 emoji. # A2UI - Agent-to-User Interface Protocol A2UI is an open-source protocol from Google that enables AI agents to generate rich, interactive user interfaces through declarative JSON rather than executable code. It solves the critical challenge of safely transmitting UI across trust boundaries in multi-agent systems - agents describe UI intent using a flat list of pre-approved components, and clients render them natively across web, mobile, and desktop. The format is optimized for LLM generation with streaming support, incremental updates, and framework-agnostic rendering. --- ## When to use this skill Trigger this skill when the user: - Wants to build an agent that generates A2UI JSON responses - Needs to set up a client renderer (Lit, React, Angular, Flutter) for A2UI - Is working with A2UI message types (`surfaceUpdate`, `createSurface`, etc.) - Wants to create or customize an A2UI component catalog - Needs to implement data binding between components and a data model - Is integrating A2UI with A2A Protocol or AG UI transport - Wants to handle client-to-server actions (button clicks, form submissions) - Is building custom components or extending the basic catalog Do NOT trigger this skill for: - General UI framework questions unrelated to agent-generated interfaces - A2A Protocol questions that don't involve UI rendering --- ## Setup & authentication ### Environment variables ```env GE...