ui

react/react

React's core repo, now shipping its own compiler — This is the official source for React, covering both the runtime and the new React Compiler (with TypeScript workflow). Useful for developers tracking React internals, contributing, or following how declarative, component-based UI building evolves across web and native.

The library for web and native user interfaces.

open-webui/open-webui

A self-hosted, provider-agnostic hub for local and cloud LLMs — Open WebUI unifies Ollama and OpenAI-compatible APIs behind one offline-capable interface, letting users mix local and cloud models freely. Granular RBAC, plugin support via Filters/Actions/Tools, and MCP/OpenAPI tool server connections make it appealing to teams needing customizable, secure self-hosted AI agents.

User-friendly AI Interface (Supports Ollama, OpenAI API, ...)

sveltejs/svelte

Svelte compiles components to lean vanilla JavaScript — Svelte turns declarative UI components into efficient JavaScript that directly updates the DOM, skipping a runtime virtual DOM. It's for web developers who want smaller bundles and simpler reactivity, maintained as an MIT-licensed open source project with a public roadmap.

web development for the rest of us

unslothai/unsloth

Unsloth ships a desktop app to run and fine-tune LLMs locally — Unsloth now offers a native desktop app for Windows, macOS and Linux that lets users run and train LLMs and diffusion models locally, supporting GGUF, MLX, Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4 and FLUX. It targets self-hosted, privacy-conscious users who want fine-tuning and inference without cloud dependency.

Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.

tamagui/tamagui

One style system, native-level performance on web and mobile — Tamagui bundles a universal style library, an optional UI kit, and an optimizing compiler that turns styled React components into plain divs with atomic CSS on web, or View with style objects on native. It targets teams sharing code between React and React Native who don't want to trade performance or code quality for that reuse.

Style React fast with 100% parity on React Native, an optional UI kit, and optimizing compiler.

ParthJadhav/app-store-screenshots

end to end app store screenshot creation using AI

ZSeven-W/openpencil

Rust-built AI vector design tool that lets agent teams design live on canvas — OpenPencil is an open-source, AI-native vector design tool built in Rust, pitched as a Pencil alternative. It ships a built-in MCP server and supports concurrent Agent Teams that turn prompts into UI directly on the live canvas, following a design-as-code approach for designers and developers using tools like Claude or Codex.

The world's first open-source AI-native vector design tool and the first to feature concurrent Agent Teams. Design-as-Code. Turn prompts into UI directly on the live canvas. A modern alternative to Pencil.

Jakubantalik/thinking-orbs

Nine canvas-based orb animations for AI agent status states — A React component library with nine hand-tuned dotted orb animations (searching, listening, weaving, etc.) for showing agent activity in chat UIs. Renders on plain 2D canvas, no WebGL, works identically across browsers, with two sizes and automatic dark/light theming.

Dotted thought-orb loading indicators for AI & agent UIs, 9 tuned types, two sizes, auto dark/light

plugin87/ux-ui-agent-skills

A design-system rulebook that makes Claude fail fast, not fake it — XIAS packs 138 design systems, 52 component specs and DTCG tokens into a dependency-free instruction layer for AI coding agents. Its standout feature is 52 objective gates—34 running in a real browser—that check contrast, ARIA, keyboard and overflow, so Claude's output either passes measurable accessibility checks or the build fails.

Turn Claude into a senior design architect: DTCG tokens, 52 components, WCAG 2.2 AA-AAA, 138 design systems, any-framework code, and 50 objective gates that fail the build instead of faking success.

enesozturk/dimensional-drawing

Turn any UI component into an Apple-style dimensional drawing — This is a Claude agent skill that measures a UI component's paddings, gaps, radii and layers, then renders them as a monochrome technical sheet in the style of Apple's accessory dimensional drawings. A companion skill animates the sheet into a short film. Useful for designers and developers who want to document component architecture without copy or content cluttering the spec.

Create Apple style dimensional drawings of UI components.

shadcn-labs/snapcn

Ready-made Remotion components for demo videos, installed via shadcn CLI — snapcn is a shadcn registry of Remotion components — text reveals, terminal sessions, device frames, AI chat streams, captions — installed with a single CLI command and owned as source in your repo. A browser-based editor can compose a full video and export an MP4 with nothing installed, useful for quick product demos.

A shadcn registry of Remotion components for product demo videos. Install a text reveal, device frame, terminal session, logo sting or a whole scene with a command.

Far-Beyond-Pulsar/WGPUI

One wgpu backend replaces GPUI's per-OS rendering paths — WGPUI forks gpui-ce (a community fork of Zed's GPUI) and swaps its Metal/Vulkan/D3D12 and Cocoa/Win32/Wayland code paths for a single wgpu+winit backend, adding WebAssembly support. The API stays compatible with gpui-ce, so existing GPUI apps need minimal changes to migrate.

A Fork of GPUI CE rewritten to use WGPU! https://pulsarnative.com/UI/