agnosticui

AgnosticUI Local (v2) is a CLI-based UI component library that copies components directly into your project. Works with AI tools, agent-driven UIs, and prompt-r

What is agnosticui?

AgnosticUI Local (v2) is a CLI-based UI component library that copies components directly into your project. Works with AI tools, agent-driven UIs, and prompt-ready workflows.

Overview of agnosticui

What it is

AgnosticUI Local (v2) is a CLI-based UI component library that copies component source directly into your project instead of leaving it inside node_modules. Components are authored as Lit Web Components — a standards-based layer over native Custom Elements — and the CLI wraps them for your framework at copy time. Supported frameworks are React, Vue, Svelte, and Lit. The project is open source under Apache-2.0. Components land under src/components/ag/ by default, a dedicated namespace that keeps AgnosticUI components separate from your own and is configurable if you prefer otherwise. The README describes the architecture as source-first, framework-agnostic, and built for both humans and AI agents. A schema-driven layer ships with three companion documents: v2/docs/sdui-architecture.md, v2/docs/llm-prompt-guide.md, and v2/docs/schema-coverage.md.

Key features

  • Source-first architecture: components are copied into your repository at install time, so you have full control over the source and native integration with your framework.
  • Framework-agnostic core: one component set wrapped for React, Vue, Svelte, or Lit, so the design system does not fork when the tech stack does.
  • AI-readable local source: tools such as Cursor, Claude, Copilot, Windsurf, Aider, Gemini, Replit, Kiro, and Antigravity read the raw files; the README states this reduces invented props and APIs and improves token efficiency.
  • AI Playbooks: Login Form, Onboarding Wizard, Discovery Dashboard, Support Center, Data Grid, Blog / Article Reader, and Landing / Marketing Page, each shipping a PROMPT-3-FRAMEWORKS.md prompt and a skin switcher.
  • Theme skins via CSS custom properties: base tokens, dark tokens, and brutalist light/dark skins, swapped by importing a token file rather than changing code.
  • Schema-driven layer with companion documents covering architecture, an LLM prompt guide, and per-component coverage.
  • Zero runtime dependency after ejection, per the README comparison table.

Use cases

  • Teams that want to read, edit, and own the component source inside their own repository.
  • Agent-driven and generative UI workflows where an LLM assembles components from local, high-fidelity context.
  • Projects that need the same component set across React, Vue, Svelte, and Lit.
  • Design systems that swap visual skins through token files instead of find-and-replace.

Getting started

bashcd your-projectnpx agnosticui-cli init --framework vue # or react, lit, sveltenpx agnosticui-cli add Button Card InputAfter running add, the terminal prints the exact import path for your project.

License and pricing

Apache-2.0. The project is open source; the README and metadata state no pricing.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/agnosticuiagnosticui.svg)](https://yunodb.com/tools/agnosticuiagnosticui?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/agnosticuiagnosticui?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/agnosticuiagnosticui.svg" alt="Featured on YunoDB" height="20"></a>