Born Accessible Ui logo

Born Accessible Ui

Community
ginader
born-accessible-ui

Use for ANY task that creates, edits, styles, refactors or reviews web UI: HTML, CSS, design tokens, Tailwind, JavaScript or TypeScript components, React, Next.js, Vue, Svelte, Angular, Astro, Lit, pages, dashboards, forms and prototypes. Enforces the three Born-Accessible protocols (plan semantics first, mathematically verified color contrast, native HTML before ARIA) and runs a Playwright + axe-core verification loop until zero automated accessibility violations remain.

Overview

Publisherginader
Repositoryborn-accessible-ui
Skill nameborn-accessible-ui
Stars
11
Forks
1
Bundled files
4
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 4 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by ginader on GitHub. Read the source before you install it.

Installation

Install the Born Accessible Ui AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/ginader/born-accessible-ui/tree/main/born-accessible-ui
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/ginader/born-accessible-ui.git /tmp/born-accessible-ui
mkdir -p .claude/skills
cp -r /tmp/born-accessible-ui/born-accessible-ui .claude/skills/born-accessible-ui
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Born Accessible Ui in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Born Accessible Ui on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Born Accessible Ui is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Born-Accessible UI

Build visually rich UI that is accessible on the first pass. Run this loop for every UI change:

Plan → Build → Verify → Fix → Report

1. Plan (Protocol 1: Silent Reasoning)

Before writing markup, write an A11Y-PLAN that covers:

  1. Landmarks: what goes into <header>, <nav aria-label>, <main>, <aside aria-label> and <footer>.
  2. Headings: one <h1> per page, then h2 and h3 without skipping levels.
  3. Controls: every interactive element, its native HTML element, its accessible name and its keyboard behavior.
  4. Contrast pairs: every text/background and UI/background color pair with its computed ratio.

Put the plan where it can't break the build: an HTML comment after <!DOCTYPE html> in a standalone page, a {/* A11Y-PLAN ... */} block in JSX/TSX, an HTML comment in Vue/Svelte/Angular templates, or your reply when you change several files.

When you build a single component, plan only its own semantics. Landmarks and the <h1> belong to the page layout, not to every component.

2. Build (Protocols 2 and 3)

  • Color tokens: define colors as custom properties and annotate every foreground token with its computed contrast ratio, rounded down: --text-muted: #cbd5e1; /* Contrast: 11.9:1 vs #111827 (AAA) */. Minimums: 4.5:1 for normal text, 3:1 for large text and for UI parts such as input borders, focus rings and icons.
  • Native HTML first: <button> for actions, <a href> for navigation, <dialog> with showModal() for modals, <details>/<summary> for disclosures, <input type="checkbox" role="switch"> for toggles, <label for> for every form control. Never put click handlers on <div> or <span>.
  • Layout: content reflows at 320 CSS px without horizontal scrolling, targets are at least 24×24 px, :focus-visible is clearly visible, and motion respects prefers-reduced-motion.

3. Verify (never skip, never estimate)

Run both checks after every UI change:

bash
# Static: recomputes every "Contrast:" comment, flags clickable <div>s and removed outlines
node <skill>/scripts/lint.mjs <changed files or src folder>

# Runtime: Playwright + axe-core (WCAG 2.2 A/AA + best practices) on the initial view and on
# hidden states (dialogs, popovers, details, tabs, menus), plus a 320px reflow check
node <skill>/scripts/audit.mjs <url-or-html-path> [more targets]
  • Audit every page or route you changed. For apps, start the dev server in the background first and pass its URLs (for example http://localhost:5173/settings). For static HTML, pass the files or a folder.
  • UI that only exists after a click, such as a conditionally rendered modal: add --click "<CSS selector of the trigger>".
  • Exit codes: 0 = no automated violations, 1 = violations (details in a11y-report/), 2 = setup or page-load problem (is the dev server running?).
  • Missing dependencies: run npm install --save-dev playwright @axe-core/playwright in the project. Only if Google Chrome isn't installed, also run npx playwright install chromium.
  • Keep a11y-report/ out of version control (add it to .gitignore).

4. Fix

  1. Read a11y-report/fix-prompt.md (or a11y-report/report.json for full details) and the lint output. Fix lint errors too: axe can't see clickable <div>s or wrong Contrast: comments. To compare against the starting point later, run the first audit with --out a11y-report/before.
  2. Fix at the source (tokens, shared components, layout rules) following references/remediation.md. Keep the visual design: change lightness rather than going monochrome, and never hide content from assistive technology to silence a rule.
  3. Run lint and audit again. Do at most 3 fix rounds (a round = fixing violations, then re-running both checks). If violations remain, stop and report them with your best explanation.
  4. Check every "needs review" item (the review lines in the audit output, also listed in fix-prompt.md and report.json). They are not failures; axe-core couldn't decide. These checks don't count toward the 3 rounds.
    • Text on gradients, images or translucent ("glass") surfaces: check the worst spot by eye, or put a solid backdrop behind the text so axe can compute it.
    • "Partially overlaps other elements", common for text inside a modal <dialog> or near positioned elements: don't change code for this. Verify the real colors in DevTools or a screenshot and record the result.

5. Report

  • Say "0 automated axe-core violations". Never claim "WCAG compliant" or "accessibility verified": automated tools catch only part of all barriers.
  • List each "needs review" item you checked by hand as manually verified, with the colors and ratio you measured.
  • List what still needs a human: a keyboard walk-through (Tab, Shift+Tab, Enter, Space, Esc, arrow keys in composite widgets), visible focus everywhere, a screen reader spot check of names, roles and states, 200% zoom, and meaningful alt text, link text and error messages.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Born Accessible Ui AI skill do?

Use for ANY task that creates, edits, styles, refactors or reviews web UI: HTML, CSS, design tokens, Tailwind, JavaScript or TypeScript components, React, Next.js, Vue, Svelte, Angular, Astro, Lit, pages, dashboards, forms and prototypes. Enforces the three Born-Accessible protocols (plan semantics first, mathematically verified color contrast, native HTML before ARIA) and runs a Playwright + axe-core verification loop until zero automated accessibility violations remain.

Why use Born Accessible Ui on TypingMind?

Because you install it once and use it with any model. Born Accessible Ui is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Born Accessible Ui in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/ginader/born-accessible-ui/tree/main/born-accessible-ui. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Born Accessible Ui?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Born Accessible Ui?

As many as you like. As long as a model supports skills, you can use Born Accessible Ui with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Born Accessible Ui AI skill free?

Yes. It is published on GitHub by ginader under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇