agentleFS
LeostrukaLeostruka/devin-bundle/skills/impeccable/SKILL.md

impeccable

Skill0 starsChanged 3 months agoMITView on GitHub

Use when the user asks for UI/UX work, wants to fix or polish a page/component, or when an interface looks bland, inconsistent, or like a default template.

168 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use cloud-architecture-diagram, shelra-workspace-quality and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MicrosoftSkill

    cloud-architecture-diagram

    Create, edit, or troubleshoot Azure-first architecture diagrams as editable draw.io files, with AWS and multicloud topology support and identity, API, and CI/CD flow diagrams. Use for architecture visualization, network boundaries, interaction flows, or missing diagram icons. Not for deploying infrastructure, live cloud discovery, security certification, or unrelated charts.

    Also in 3 projects that use this

  • ShelraSkill

    shelra-workspace-quality

    Audit or refine ShelraCode's main OpenTUI coding workspace when chat activity, live status, plan/context sidebar, agent visibility, or verification UX must match a supplied visual specification. Do not use for onboarding, settings, or unrelated screens.

    Also in 3 projects that use this

  • owshq-mecSkill

    layout

    Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hierarchy. Use when the user mentions layout feeling off, spacing issues, visual hierarchy, crowded UI, alignment problems, or wanting better composition.

    Also in 8 projects that use this

  • FldicoahkiinSkill

    contribute-catalog

    Author a new HyperFrames registry block (caption style, VFX block, transition, lower third) or component (text effect, overlay, snippet) and ship it as an upstream PR to the hyperframes repo. Use ONLY when the user wants to CONTRIBUTE to the public catalog — for in-project caption/transition authoring use the `hyperframes` skill, for installing existing registry items use the `hyperframes-registry` skill.

    Also in 3 projects that use this

  • HK-hubSkill

    overdrive

    Pushes interfaces past conventional limits with technically ambitious implementations — shaders, spring physics, scroll-driven reveals, 60fps animations. Use when the user wants to wow, impress, go all-out, or make something that feels extraordinary.

    Also in 11 projects that use this

  • HK-hubSkill

    bolder

    Amplify safe or boring designs to make them more visually interesting and stimulating. Increases impact while maintaining usability. Use when the user says the design looks bland, generic, too safe, lacks personality, or wants more visual impact and character.

    Also in 12 projects that use this

  • +3 more from HK-hub

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. 0 are common, 9 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 167 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Not for backend-only or non-UI tasks.

    line 16
  2. Very specific:

    Gather design context first: If the project lacks PRODUCT.md and DESIGN.md (or .impeccable.md) with the required context, ask the user or …

    line 20
  3. Distinctive:

    Audit before declaring done: Check against common slop patterns:

    line 31
  4. Distinctive:

    Use this loop when the user says something like "fix the design" without naming a specific command.

    line 72
  5. Very specific:

    Run the page in browser_preview.

    line 75
  6. Distinctive:

    Check for: accessibility labels, focus rings, color contrast, motion preference, no-JS fallbacks, layout overflow, and component misuse.

    line 76
  7. Distinctive:

    Look for slop and score the page.

    line 80
  8. Distinctive:

    After each pass that produces code changes:

    line 97
  9. Very specific:

    Run php artisan test if PHP/Blade changed.

    line 98
  10. Very specific:

    Run npm run build if CSS/JS changed.

    line 99
  11. Distinctive:

    Commit with a focused message.

    line 100
  12. Distinctive:

    Prefer real data and concrete labels over lorem ipsum.

    line 113
  13. Very specific:

    Use browser_preview after significant UI changes.

    line 114
  14. Very specific:

    If the user names a UI framework (React, Vue, Svelte, Tailwind, shadcn, etc.), invoke context7 before proposing implementation details.

    line 115
  15. Very specific:

    If a local npx impeccable CLI is installed and trusted, prefer its detector output for audit and critique.

    line 116
  16. Distinctive:

    Keep files small and focused.

    line 117

“Common” is measured: kept by at least 6 in 10 of the 167 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Impeccable

Design vocabulary for production-grade frontend interfaces. Adapted from https://impeccable.style/ by Paul Bakaus (Apache 2.0).

When to use

  • User asks for UI/UX work: landing page, dashboard, component, app shell, form, settings, onboarding, empty state.
  • Existing UI looks bland, loud, inconsistent, or like a default template.
  • Need to audit, polish, or iterate a frontend surface.
  • Not for backend-only or non-UI tasks.

Core protocol

  1. Gather design context first. If the project lacks PRODUCT.md and DESIGN.md (or .impeccable.md) with the required context, ask the user or infer from existing code. Required context:
    • Target audience and use cases.
    • Brand personality/tone (bolder vs quieter, warm vs cold, playful vs serious).
    • Anti-references (styles to avoid).
    • Color, typography, and component constraints.
    • Surface type: brand (marketing/landing/portfolio) or product (app/dashboard/tool).
  2. Pick one command that matches the goal. Do not chain passes unless the user asks.
  3. Implement real working code with attention to spacing, hierarchy, typography, color, motion, and edge cases. Use browser_preview to verify visual output when possible.
  4. Audit before declaring done. Check against common slop patterns:
    • Generic gradients, purple-to-blue, beige body background, cards nested in cards.
    • Rounded-square icon tiles above every heading.
    • Gray text on colored backgrounds, Inter for everything.
    • Vague headlines like "Unlock your potential".
    • Equal visual weight on unrelated metrics.

Push vs Pull for Design Patterns

  • Push (to critique/audit): the reviewer or audit pass receives the design context — PRODUCT.md, DESIGN.md or .impeccable.md, anti-references, brand tone, surface type, and the slop checklist. Push these constraints before the critique so the reviewer can verify the surface against them.
  • Pull (by implementer): the implementer reaches for the right impeccable command (shape, layout, harden, etc.) and the slop checklist before writing UI, instead of waiting for the critique to catch generic design. Pull the relevant command for the pass you are about to run, then push the resulting constraints into the next review.

Command vocabulary

CommandUse when
impeccableGeneral design request; pick next step.
shapePlan UX/UI before writing code.
polishMeticulous final pass between good and great.
critiqueDesign review with scoring and detection.
auditTechnical quality check (layout, performance, accessibility).
distillRuthless subtraction; strip to essence.
clarifyRewrite confusing UX copy.
layoutFix spacing, rhythm, alignment.
typesetFix typography that feels generic or accidental.
colorizeAdd strategic color without going garish.
animatePurposeful motion that conveys state.
bolderPush safe designs toward impact.
quieterTone down designs that are shouting.
delightSmall moments of personality.
adaptResponsive or cross-context adaptation.
hardenProduction-ready edge cases, i18n, errors, overflow.
onboardFirst-run experiences, empty states, paths to value.
optimizeUI performance (LCP, bundle size).
extractPull reusable components/tokens into a design system.
documentGenerate DESIGN.md from existing code.
liveIterate UI in the browser with variants.
initOne-time setup of PRODUCT.md and DESIGN.md.

Frontend loop

Use this loop when the user says something like "fix the design" without naming a specific command. Run the passes in order. Stop when the result is good enough; otherwise write a short plan, fix, then move to the next pass.

  1. audit — Technical quality check.
    • Run the page in browser_preview.
    • Check for: accessibility labels, focus rings, color contrast, motion preference, no-JS fallbacks, layout overflow, and component misuse.
    • Output: findings list in .devin/plans/impeccable-<surface>-audit.md.
  2. critique — Design review.
    • Look for slop and score the page. List the top 3 issues.
    • Output: .devin/plans/impeccable-<surface>-critique.md.
  3. distill (optional) — Strip non-essential elements that do not support the task.
  4. layout — Fix spacing, rhythm, alignment, and responsive breakpoints.
  5. typeset — Check hierarchy, line-height, font weights, and label casing.
  6. colorize — Verify all color uses map to the project theme tokens (primary, on-primary, surface, etc.). Fix mismatched tokens like bg-primary-container text-on-primary.
  7. harden — Test reduced motion, keyboard navigation, screen-reader labels, error states, long text, small viewports, and dark mode.
  8. polish — Final pass on shadows, transitions, border radius consistency, and icon sizing.

Output rule

After each pass that produces code changes:

  • Run php artisan test if PHP/Blade changed.
  • Run npm run build if CSS/JS changed.
  • Commit with a focused message.

Verification checklist

  • No generic template aesthetics.
  • Color tokens are correct for each surface.
  • Focus states are visible.
  • Motion respects prefers-reduced-motion.
  • No-JS fallback works where applicable.
  • Tests and build pass.

Implementation notes

  • Prefer real data and concrete labels over lorem ipsum.
  • Use browser_preview after significant UI changes.
  • If the user names a UI framework (React, Vue, Svelte, Tailwind, shadcn, etc.), invoke context7 before proposing implementation details.
  • If a local npx impeccable CLI is installed and trusted, prefer its detector output for audit and critique; otherwise apply the principles above.
  • Keep files small and focused; do not let design context leak into backend logic.

Cross-skills

  • a11y-audit — when polish touches interaction/markup, verify WCAG (contrast, focus, labels) before shipping.
  • creative-engineering — when the user names a visual style/archetype (CRT, dither, brutalist, vaporwave…), load its technical anatomy instead of improvising.

References

  • https://impeccable.style/docs/ — full command reference.
  • https://impeccable.style/slop/ — anti-pattern gallery.
  • https://impeccable.style/designing/ — design principles.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.