UX/UI Agent Skills is a set of instructions, design tokens, reusable components, accessibility rules, and design-system references that guide Claude in producing interface designs and code. It is intended for developers and designers who want framework-independent, token-based UI work with accessibility checks. The catalogue entries package these UX, UI, component, and validation workflows as Claude Code skills and related add-ons.
Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/plugin87/ux-ui-agent-skillsnpx agentmods add commands/plugin87/ux-ui-agent-skills/gateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/plugin87/ux-ui-agent-skills/gate)<a href="https://agentmods.dev/commands/plugin87/ux-ui-agent-skills/gate"><img src="https://agentmods.dev/badge/commands/plugin87/ux-ui-agent-skills/gate/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/plugin87/ux-ui-agent-skills/gate"><img src="https://agentmods.dev/badge/commands/plugin87/ux-ui-agent-skills/gate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00025 | $0.00603 |
| Opus 5.5 | $0.00010 | $0.00241 |
| Sonnet 5.5 | $0.00005 | $0.00121 |
| Haiku 4.5 | $0.00003 | $0.00060 |
Grade A, and why
gate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run the checks below in order, against this project. Report the real output of
each one, then a single N/N line. It is all or nothing: if a check fails, fix
it and re-run rather than reporting a partial pass.
<harness> means the component's *.states.html file, or the directory
src/components/ to sweep all of them.
One-time setup: the render gates drive headless Chrome, so a fresh project needs
npm i -D playwright once. Without it those scripts print SKIPPED and exit 0,
which looks like a pass and is not one.
Theme (the source of truth)
python3 scripts/validate_tokens.py design-tokens.json
python3 scripts/validate_contrast.py design-tokens.json
node scripts/build_tokens.mjs --in design-tokens.json --out src/theme.css
python3 scripts/validate_theme_refs.py src/theme.css src
python3 scripts/lint_hardcodes.py src
python3 scripts/check_no_emoji.py src public
Rendered UI (every component harness)
node scripts/verify_states.mjs src/components/<harness>
node scripts/verify_states.mjs --dark src/components/<harness>
node scripts/axe_audit.mjs src/components/<harness>
node scripts/verify_responsive.mjs src/components
node scripts/verify_responsive.mjs src/components --scale=1.25
node scripts/verify_target_size.mjs src/components
node scripts/verify_keyboard.mjs src/components
node scripts/verify_reduced_motion.mjs src/components
node scripts/verify_overflow.mjs src/components
node scripts/lint_intent.mjs src/components
Modals and drawers add:
node scripts/verify_focustrap.mjs src/components/<harness> --open="#openBtn"
Then look
The gates prove contrast, roles, sizes, and overflow. They do not prove the UI works. Screenshot the harness in light and dark, click every control, and confirm the state actually changed. A checkbox that passes axe and does not toggle is still broken.
Scope, stated honestly
This proves objective correctness: token consistency, WCAG 2.2 AA on a real render in both themes, keyboard operability, target size, reduced motion, responsive behaviour, no hardcoded values, no emoji. It does not prove taste. Never report a number a gate did not just produce.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago Changed · +39 lines · +1 tokens per session ddcf6b314c91
- 1mo ago First seen · 26 lines · 24 tokens per session scan A 992e800de85d
gate is a command published in the GitHub repository plugin87/ux-ui-agent-skills (2,012 stars, last pushed 7d ago), licensed MIT. It adds 25 tokens to every session and 603 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
test-strategy
Write a test strategy as Markdown in the user's repository. Produce designed HTML or PDF only when asked.
write-tests
Write tests for: $ARGUMENTS.
apex
APEX Methodology - The systematic Analyze-Plan-Execute-eLicit-eXamine approach for intelligent development. Reduces hallucination and defect risk through mandatory parallel research, self-review, and validation gates.
adr
Write an architecture decision record as Markdown in the user's repository. Use when invoked as /document-design-system:adr. Produce designed HTML or PDF only when asked.
support-model
Write a support model as Markdown in the user's repository. Produce designed HTML or PDF only when asked.
construct
Run the CONSTRUCT stage of the AI Workflow Playbooks pipeline.