Skip to main content

Scaffolding CLI for Agent UI Starter Pack

Project description

Agent UI Starter Pack

A high-performance, zero-dependency A2UI (Agentic-Adaptive User Interface) renderer built with React, TypeScript, and Vite.

🚀 Overview

This project provides a foundational implementation of the A2UI protocol, allowing AI agents to dynamically generate and update user interfaces using a declarative JSON schema. It is designed to be lightweight, premium, and easy to extend.

✨ Features

  • Zero-Dependency Renderer: Recursive rendering engine that handles complex A2UI JSON trees without external SDKs.
  • Multi-Platform Support: Built-in logic optimized for React, Lit (Web Components), and Flutter (GenUI SDK).
  • Premium Design System: Dark-mode aesthetic with glassmorphism, smooth transitions, and professional typography (Inter).
  • Protocol Ready: Native support for A2UI (Agent-to-User) and A2A (Agent-to-Agent) standards.
  • Agent Pulse: Built-in "Agent Online" status indicator with CSS animations.

🛠️ Getting Started

You can scaffold a new A2UI project in seconds without installing anything using uvx. Once the package is published to PyPI, the command is simply:

# Basic setup
uvx agentui-starter-pack create my-ui

# High-fidelity setup with CopilotKit/AGUI
uvx agentui-starter-pack create my-ui --ui agui --copilotkit

[!NOTE] Until the package is published to PyPI, use the --from flag to point to the GitHub repository: uvx --from git+https://github.com/enriquekalven/agent-ui-starter-pack.git agentui-starter-pack create my-ui

Prerequisites

  • Node.js (v18+)
  • Python (v3.10+) for the CLI
  • git
# Install dependencies
npm install

Development

# Start the Vite development server
npm run dev

The application will be available at http://localhost:5173.

🚀 Deployment

The project includes one-line deployment commands via the Makefile.

Platform Command
Cloud Run make deploy-cloud-run
Firebase make deploy-firebase
GKE make deploy-gke

For more options and configuration details, see the Deployment Guide.

🏗️ Architecture & Frameworks

Web Components (Lit)

A2UI's core repository is built with Lit. For the "Full Control" path, you can use the native Lit renderer to create standard Web Components that work in any framework (including Angular).

Flutter (GenUI SDK)

A2UI is designed for developers building custom surfaces across Web, Mobile, and Flutter. The GenUI SDK pattern allows for "Native-First" rendering where the agent sends blueprints that are realized using high-performance Flutter widgets.

AG UI (CopilotKit)

For high-end React implementations, we recommend CopilotKit (formerly AG UI). It provides a full-stack starter designed for A2A and A2UI, handling the streaming "pipes" so you can focus on building beautiful surfaces.

📡 A2A Protocol

The Agent-to-Agent (A2A) protocol enables agents to communicate over a network. See the A2A_GUIDE.md for details on exposing agents as services and orchestrating multi-agent systems.

📜 A2UI Protocol

The renderer expects an A2UISurface object with the following structure:

{
  "surfaceId": "unique-id",
  "content": [
    {
      "type": "Text",
      "props": { "text": "Hello World", "variant": "h1" }
    },
    {
      "type": "Card",
      "props": { "title": "System Meta" },
      "children": [...]
    }
  ]
}

📄 License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agentui_starter_pack-0.1.1.tar.gz (14.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agentui_starter_pack-0.1.1-py3-none-any.whl (4.9 kB view details)

Uploaded Python 3

File details

Details for the file agentui_starter_pack-0.1.1.tar.gz.

File metadata

  • Download URL: agentui_starter_pack-0.1.1.tar.gz
  • Upload date:
  • Size: 14.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.2

File hashes

Hashes for agentui_starter_pack-0.1.1.tar.gz
Algorithm Hash digest
SHA256 425809562af15802b67a59988eba757fc18201f2a78ec512a2849c2ea0e10ea5
MD5 6864f0cc13c07d329d79e805ef84a145
BLAKE2b-256 34991a2b7086338a40ac394cff3fc6bd3d41c0c478fc636985b2fb65fbaa2f39

See more details on using hashes here.

File details

Details for the file agentui_starter_pack-0.1.1-py3-none-any.whl.

File metadata

File hashes

Hashes for agentui_starter_pack-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 89e8c1a7f8223e6e26b7b18d14c5dc93f9e767517d081450dbe8e23454697503
MD5 fe699793599ca97024848a7c8ebb0bc9
BLAKE2b-256 c3dbcc1b835afc5672a76e8ee24de87f9c730d82447490276e2d3b4881367f80

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page