google-labs-code/stitch-sdk

Generate UI screens from text prompts and extract their HTML and screenshots programmatically.

What it solves

It allows developers to programmatically generate, edit, and iterate on UI screens using text prompts. Instead of manually designing interfaces, users can describe a screen in plain language and receive the corresponding HTML and screenshots.

How it works

The SDK connects to a backend service that transforms text descriptions into UI designs. It organizes work into "projects," where users can generate new screens, refine existing ones through editing prompts, or create multiple design variants (adjusting layout, color schemes, or fonts). It also supports the creation of design systems to maintain visual consistency across screens.

Who it’s for

  • Frontend Developers: To rapidly prototype UI layouts and get starter HTML.
  • AI Agent Developers: To build autonomous designers that can create and modify interfaces using the provided tool integrations for Vercel AI SDK and Google ADK.
  • Product Designers: To quickly explore different visual directions for a screen.

Highlights

  • Text-to-UI Generation: Create full screens from simple prompts.
  • UI Iteration: Edit existing screens or generate multiple variants with specific creative ranges (Refine, Explore, Reimagine).
  • Agent-Ready: Includes built-in support for MCP (Model Context Protocol) tools, making it easy to integrate with LLM agents.
  • Design System Support: Create and apply consistent visual themes across multiple screens.
  • Multi-format Output: Provides direct download URLs for both the HTML code and a screenshot image of the generated UI.

Related

  • Project
    google-labs-code/stitch-skills

    A collection of agent skills and plugins for Google Stitch that enables AI coding agents to automate design generation, design system management, and conversion of designs to frontend code.

  • Project
    vercel/chat

    Chat SDK is a TypeScript library that lets you write a single bot and run it on many chat platforms (Slack, Teams, Discord, etc.) via plug‑in adapters, with built‑in support for AI streaming, rich UI cards, and state management.

  • Project
    oil-oil/draw-ui

    draw‑ui is an AI‑powered skill that helps an LLM agent design a UI from a text brief or existing screenshot and optionally output ready‑to‑use HTML/CSS or WeChat mini‑program code.

  • Project
    Leonxlnx/taste-skill

    Taste Skill is a set of portable Agent Skills that guide AI coding assistants to produce cleaner, design‑focused front‑end code and optional design mock‑ups. Install via `npx skills add …`, choose a skill (e.g., `design‑taste‑frontend`), and adjust three numeric dials to control layout variance, motion intensity, and visual density.

  • Project
    strands-agents/harness-sdk

    An open-source SDK for building and running AI agents in Python and TypeScript, providing a model-agnostic framework for managing agent loops, tools, and memory.