Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add PromptPartner/agentsmith --skill design-reviewgit clone --depth 1 https://github.com/PromptPartner/agentsmithWrote 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/skills/promptpartner/agentsmith/design-review)<a href="https://agentmods.dev/skills/promptpartner/agentsmith/design-review"><img src="https://agentmods.dev/badge/skills/promptpartner/agentsmith/design-review.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00061 | $0.00995 |
| Opus 5 | $0.00030 | $0.00498 |
| Sonnet 5 | $0.00012 | $0.00199 |
| Haiku 4.5 | $0.00006 | $0.00100 |
Grade A, and why
design-review 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 8d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design Review — Facet UI
verify.sh proves a component builds, types, tests, and passes axe. It cannot prove the component
matches the design system — "is this on-system?" is a judgment call, not a grep (see
docs/11). This skill is that judgment pass, as a checklist,
so a component doesn't ship looking almost-right. It loads only when you're reviewing UI — everyday
work doesn't pay for it.
When this fires
You type /design-review, or you've added/restyled a component and are about to open or merge the PR.
Before you start
Open DESIGN.md and keep it beside the diff. Every check below is "does the
component agree with DESIGN.md?" — if you're guessing, you haven't read it recently enough.
Checklist
Walk these in order. Stop at the first miss, fix it, restart — an off-system component is not "done."
- Tokens, not literals. No raw hex,
pxfont sizes, or magic radii in the component — every value resolves to a token insrc/tokens/that mirrorsDESIGN.md(color.primary,radius.md,type.body,space.lg). A literal#5e6ad2or13pxis a miss even if it looks identical, because the next token change won't reach it. - All states present and on-system. Default, hover, active,
:focus-visible, and disabled — each matchingDESIGN.md(e.g. a primary button: hover →primary-hover#828fff, focus → a 2pxprimary-focusring). A missing state is the most common miss; check every interactive part. - Accessible. Keyboard-operable, correct ARIA role/name, a visible focus ring (never
outline: nonewithout an equal replacement), and contrast ≥ 4.5:1 for body text (≥ 3:1 for large / UI).test-storybook's axe pass is necessary but not sufficient — eyeball the focus ring. - Type and spacing scale. Sizes, weights, line-heights, and letter-spacing come from the type
scale; padding/gaps come from the spacing scale. No one-off
600weight whereDESIGN.mdsays500, no10pxgap outside the 4px grid. - Right surface and elevation. Backgrounds use the surface ladder (
canvas → surface-1 → …) and depth is carried by surface + hairline border perDESIGN.md— not an invented drop shadow or a skipped surface level. - Responsive. The component reflows per the breakpoints and holds touch-target minimums (CTAs ≥ 40px, ≥ 44px on touch). Check the smallest supported width, not just desktop.
DESIGN.mdmoved with the code (R6). If you added a new component pattern or variant, its entry exists inDESIGN.md'scomponents:section in this same PR. If the pattern you needed wasn't inDESIGN.md, it should have been added there first — fix the order, don't paper over it.- Story + test cover the above. The
.stories.tsxshows every state (so the a11y/interaction runner and a reviewer both see them); the.test.tsxasserts them. A state with no story is a state no one reviews.
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.
- 8d ago First seen · 61 lines · 61 tokens per session scan A 4f40ade9bc3d
design-review is a skill published in the GitHub repository PromptPartner/agentsmith (185 stars, last pushed 3d ago), licensed MIT. It adds 61 tokens to every session and 995 once invoked, about $0.0003 per session on Opus 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 skills, from other repositories
meta-design-composable-components
Composable component APIs — parts, state, polymorphism.
desktop-ui-electron
Frameless windows, custom title bars, tray, menus, dock badges, vibrancy, kiosk mode, window state persistence.
mobile-ui-components-react-native-paper
React Native Paper v5+ - Material Design 3 theming, PaperProvider, key components, dynamic color, accessibility, tree-shaking, custom fonts, navigation integration.
web-accessibility-web-accessibility
WCAG, ARIA, keyboard navigation.
web-maps-mapbox
Mapbox GL JS interactive vector maps. Use when building a map with sources and layers, expressions, clustering, 3D terrain, geocoding or directions.
web-animation-css-animations
CSS animation patterns - transitions, keyframes, scroll-driven timelines, @property, compositor-friendly properties, prefers-reduced-motion.