Impeccable design
Skill All-The-Vibes/ATV-PaperBoard/adapters/copilot-cli/_dist/skills/impeccable-design
Bind agent-authored `.DESIGN.md` files to paperboard's design doctrine. Loads the impeccable anti-slop methodology (vendored, Apache 2.0, pinned to commit `4af581e23f17d112d8f9d6b7a5b7ff37823494e1`) and the binding rules in `core/designs/DESIGN-AUTHORITY.md`. Read these BEFORE writing or editing any `.DESIGN.md` file.From its SKILL.md
npx -y skills add All-The-Vibes/ATV-PaperBoard --skill impeccable-designAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
- runs commandsInstructs the agent to run 2 commands, including `pytest -q` and 1 more.
SKILL.md
5.9 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it
impeccable-design
Paperboard's design authority skill. It does not run a CLI — it loads doctrine
into the agent's context so that every authored or modified .DESIGN.md file
respects the same anti-slop rules. Methodology is imported from
pbakaus/impeccable; aesthetic tokens
remain paperboard's own.
When to invoke
- The user asks you to create, edit, or audit a
.DESIGN.mdfile (a design tier, a starter, or the defaultpaperboard.DESIGN.md). - The user asks for a new design tier, a new starter pack entry, or a redesign of an existing tier.
- You are reviewing a PR that touches anything under
core/designs/. - A
.DESIGN.mdlint fails and you need to reason about why a token or rule is banned.
Do not invoke for plain HTML/CSS work that does not touch a .DESIGN.md
contract, or for documentation prose unrelated to design tokens.
Doctrine sources
Read these files, in order, before authoring or modifying a .DESIGN.md:
core/designs/DESIGN-AUTHORITY.md— the binding rules. The single source of truth for what is allowed in any paperboard.DESIGN.md.core/designs/impeccable-context/typography.md— type scale, weights, tracking, mono-vs-serif decisions.core/designs/impeccable-context/color-and-contrast.md— palette discipline, accent budget, tinted neutrals, contrast floor.core/designs/impeccable-context/spatial-design.md— spacing scale, border discipline, depth from negative space (not shadows).core/designs/impeccable-context/motion-design.md— easing curves (cubic-bezier(0.16, 1, 0.3, 1)only; no bounce), durations,prefers-reduced-motion.core/designs/impeccable-context/interaction-design.md— affordance, state, focus, hit-target rules.core/designs/impeccable-context/responsive-design.md— breakpoint philosophy, fluid type, container query intent.core/designs/impeccable-context/ux-writing.md— voice, hedging bans, expert-decisive tone.
Provenance: vendored verbatim under Apache 2.0 from pbakaus/impeccable at
commit 4af581e23f17d112d8f9d6b7a5b7ff37823494e1. See
core/designs/impeccable-context/UPSTREAM.md and NOTICE.md.
Absolute bans (non-negotiable)
These are restated from DESIGN-AUTHORITY.md so the skill is self-contained.
Every .DESIGN.md Don't list must enforce all of them:
- Glassmorphism — no
backdrop-filter: blur, no frosted-translucent surfaces, no glow-on-blur cards. - Gradient text — no
-webkit-background-clip: textwith a gradient fill. - Side-stripe borders — no
border-left/border-rightgreater than 1px used as a colored stripe on cards, list items, callouts, or alerts. - Nested card-in-card — no card containers visually nested inside other cards. Flatten the hierarchy; depth comes from negative space.
- Generic AI emoji decoration — no ✨ 🚀 ⚡ 🎯 🔥 💎 or comparable cliché
glyphs used decoratively. Status dots, mono
·, and typographic emphasis are the only allowed ornaments. - Identical-card feature grids — no
repeat(auto-fit, minmax(280px, 1fr))endlessly tiling same-shaped icon-heading-text cards. - Hero-metric SaaS layouts — no big-number-with-tiny-label-and-gradient template.
- Bounce / elastic easing — animations decelerate via expo-out only.
- Pure
#000/#fff— always tinted neutrals (paperboard ships#08090A/#F7F8F8). - Hedging prose — "maybe consider," "could be helpful," "might want to" are banned. Design rules are decisive.
If a request asks for any of the above, refuse and explain which ban applies.
How to use this when authoring or modifying a .DESIGN.md file
- Read
core/designs/DESIGN-AUTHORITY.mdin full before writing any token or rule. Then read the relevant impeccable-context file for the dimension you are touching (typography, color, spatial, motion, etc.). - Audit the existing file (if any) against the bans above. List every violation found. Do not silently fix — list, then fix.
- Write or edit tokens in paperboard's voice: cold, dark, technical, expert-decisive. No hedging. Every Don't is a hard, enforceable ban.
- Add or update the audit comment at the very top of the file, in HTML
comment form:
<!-- YYYY-MM-DD: audited against impeccable doctrine; <N> violations corrected: <list> -->or<!-- YYYY-MM-DD: audited against impeccable doctrine; no violations found -->. - Verify by running
pytest -qand (when Node.js is available)npm run lint:artifacts. Neither must regress.
Methodology vs. aesthetic — the load-bearing distinction
Paperboard adopts impeccable's methodology (how design rules are authored and enforced), not its aesthetic (the warm Cormorant Garamond editorial palette). Impeccable's own tokens — Cormorant Garamond, Warm Ash Cream, Editorial Magenta — are not adopted by any paperboard tier.
Paperboard's canonical aesthetic remains cold, dark, technical:
#08090Anear-black canvas;#F7F8F8tinted off-white.- Geist Mono eyebrows in UPPERCASE with
0.12emtracking. - Indigo / rose accents capped at ≤ 15% of visible surface area.
- 1px hairline borders; depth via tonal layering, not drop shadows.
- Numbered sections (
00,01,02) with tightened heading tracking (-0.02emto-0.04em).
If you find yourself writing tokens that look "warm editorial," stop — you have crossed from methodology into aesthetic import. Roll back.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most design frontend skills give in ~1.5k tokens
Counted across 1,179 of the 2,086 authors here whose files we hold, read 2026-09-06
- Commit to a bold aesthetic directionin 31 of 1179, across 24 files
- Prefer component composition over inheritancein 28 of 1179, across 14 files
- Animate only transform and opacity propertiesin 27 of 1179, across 22 files
- Memoize expensive computations with useMemoin 26 of 1179, across 13 files
- Use semantic HTML elementsin 24 of 1179, across 23 files
- Virtualize long lists for performancein 21 of 1179, across 10 files
- Use CSS variables for design tokensin 20 of 1179, across 14 files
- Implement loading, empty, and error statesin 20 of 1179
- Lazy load heavy components with Suspensein 19 of 1179, across 8 files
- Respect prefers-reduced-motion media queriesin 18 of 1179, across 10 files
- Prioritize CSS-only animations for HTMLin 18 of 1179, across 16 files
- Use compound components for related UI elementsin 18 of 1179, across 7 files
Said here and by no other author read
- read design authority files before authoring any design file
- audit existing design files against absolute bans
- write design tokens in a cold and decisive tone
- add an audit comment to the top of design files
- verify changes by running test and lint suites
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.