Brandkit
WCAG-verified DTCG design-token foundation (color, type, spacing, theme.css) from a brief
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 8ebac21
Located SKILL.md at .claude/skills/brandkit/; the dir ships SKILL.md only, its referenced scripts/ + taste/ + CLAUDE.md live at repo root (all HTTP 200). For OUTPUT I invented a brief (fintech savings app, teal, "trustworthy") and wrote two DTCG colors.json: a naive baseline (action=#14b8a6 with white text, text.secondary=gray-400) and a skill-following one (action=teal-700, gray-600 secondary, designed dark overrides). Ran the repo's own scripts/validate_contrast.py against each: baseline = 7 required WCAG pairs FAIL (exit 1: link 2.49:1, on-action 2.49:1, secondary 2.54:1, border 2.54:1, plus 3 in dark); skill-following = all 12 required pairs PASS in light AND dark (exit 0). Installed into a temp HOME and confirmed validate_contrast.py and validate_tokens.py both exit 0 from the skill dir. SKILL.md is honest ("feel still benefits from human review"); security clean.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 4/5
What Brandkit does
Generates a 3-tier DTCG design-token system (primitive to semantic to component) from a brand brief: OKLCH color ramps, type/spacing/radius/shadow/motion scales, light and dark, plus a single theme.css. Ships Python validators that check token aliases resolve and that required text/action/border pairs meet WCAG 2.2 AA in both light and dark. Triggers when the user wants a from-scratch brand or design foundation, a new palette and type system, or a themeable token kit for a product.
How to install Brandkit
git clone --depth 1 https://github.com/plugin87/ux-ui-agent-skills.git /tmp/brandkit-src
mkdir -p ~/.claude/skills/brandkit
cp -R /tmp/brandkit-src/.claude/skills/brandkit/. ~/.claude/skills/brandkit/
cp -R /tmp/brandkit-src/scripts /tmp/brandkit-src/taste /tmp/brandkit-src/tokens /tmp/brandkit-src/CLAUDE.md ~/.claude/skills/brandkit/
# The brandkit dir upstream is SKILL.md only; its body references scripts/, taste/, tokens/ and
# CLAUDE.md that live at the REPO ROOT, so the two extra cp lines co-locate them into the skill dir.
# Verifiers are pure-stdlib Python3 (no pip deps): run from ~/.claude/skills/brandkit with
# python3 scripts/validate_tokens.py && python3 scripts/validate_contrast.py && python3 scripts/validate_theme_refs.py
# Alternative whole-repo install (npx scaffolder, lays down the same files in a project): npx ux-ui-agent-skills
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Brandkit
-
/brandkitWCAG-verified DTCG design-token foundation (color, type, spacing, theme.css) from a brief
It also activates on plain-language prompts like these:
-
Build a brand design system from this brief -
Generate a light and dark theme token kit -
Create a WCAG-verified color and type system
Frequently asked questions
- Is the Brandkit skill free?
- Yes. The skill itself is free from plugin87/ux-ui-agent-skills. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Brandkit work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Tested · Works. Located SKILL.md at .claude/skills/brandkit/; the dir ships SKILL.md only, its referenced scripts/ + taste/ + CLAUDE.md live at repo root (all HTTP 200). For OUTPUT I invented a brief (fintech savings app, teal, "trustworthy") and wrote two DTCG colors.json: a naive baseline (action=#14b8a6 with white text, text.secondary=gray-400) and a skill-following one (action=teal-700, gray-600 secondary, designed dark overrides). Ran the repo's own scripts/validate_contrast.py against each: baseline = 7 required WCAG pairs FAIL (exit 1: link 2.49:1, on-action 2.49:1, secondary 2.54:1, border 2.54:1, plus 3 in dark); skill-following = all 12 required pairs PASS in light AND dark (exit 0). Installed into a temp HOME and confirmed validate_contrast.py and validate_tokens.py both exit 0 from the skill dir. SKILL.md is honest ("feel still benefits from human review"); security clean.
- What is the Brandkit SkillProof Score?
- 8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 4/5.
- How do I install Brandkit?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Brandkit with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.