agentleFS
kynn-devkynn-dev/agent-skills/.agents/skills/frontend-design/SKILL.md

frontend-design

Skill1 starsChanged 46 days agoMITView on GitHub

Use when designing, implementing, or reviewing substantive UI, interaction, layout, responsive behavior, visual hierarchy, or front-end motion; do not use for copy-only work, brand positioning without evidence, backend/data architecture, or generic visual advice detached from a real product surface.

187 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use instantly-campaign, albert-dm 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.
  • yawningwhaleAISkill

    instantly-campaign

    Create a new Instantly cold-email campaign in the proven 5-email structure of the user's top-performing campaign (~16% reply rate). Use when the user wants to launch a new outbound campaign, write a cold email sequence, create an Instantly campaign, draft outreach for a new service or niche, or set up a sales sequence.

    Also in 6 projects that use this

  • yawningwhaleAISkill

    albert-dm

    Reply to DMs and messages the way Albert would. Use when the user asks "how would Albert reply to this", pastes a conversation and wants an Albert-style response, or asks you to draft/write a DM in Albert's voice. Albert sells AI voice agents / automations to local service businesses (Go High Level, lead follow-up, bookings).

    Also in 5 projects that use this

  • naman10parikhSkill

    ecosystem-reference

    Quick-access directory of Claude Code ecosystem repos, skills, plugins, MCPs, and patterns. Use when improving the harness.

    Also in 5 projects that use this

  • naman10parikhSkill

    fractal-delegation

    Also in 5 projects that use this

  • julianoczkowskiSkill

    brief-to-tasks

    Break a design brief into an ordered checklist of independently buildable tasks using vertical slices. Saves as a markdown checklist. Use when user wants to break down work, create tasks from a brief, plan implementation order, or mentions "tasks" or "breakdown".

    Also in 7 projects that use this

  • SeifBenayedSkill

    file-reading

    Use this skill when a file has been uploaded but its content is NOT in your context — only its path at /mnt/user-data/uploads/ is listed in an uploaded_files block. This skill is a router: it tells you which tool to use for each file type (pdf, docx, xlsx, csv, json, images, archives, ebooks) so you read the right amount the right way instead of blindly running cat on a binary. Triggers: any mention of /mnt/user-data/uploads/, an uploaded_files section, a file_path tag, or a user asking about an uploaded file you have not yet read. Do NOT use this skill if the file content is already visible in your context inside a documents block — you already have it.

    Also in 6 projects that use this

  • +6 more from naman10parikh

The file

Show its rules and text

We split this skill into the rules it gives an agent

23 rules. 0 are common, 21 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 181 edited versions we read keep these.

  • 21

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the target route/surface, product task, repository/component system, existing tokens/assets/identity evidence, content and data states, …

    line 18
  2. Distinctive:

    Use this skill when the request asks to design, implement, revise, or review:

    line 30
  3. Distinctive:

    Do not use it for:

    line 39
  4. Distinctive:

    If the request mixes design with another domain, split the work and route each bounded part to the appropriate skill.

    line 47
  5. Distinctive:

    Before writing UI code, establish five things:

    line 51
  6. Distinctive:

    If the current identity is incomplete, mark the gap and make reversible decisions rather than silently inventing brand rules.

    line 61
  7. Distinctive:

    Do not turn a heuristic score into false certainty.

    line 75
  8. Distinctive:

    Use semantic HTML and the project's established framework/component conventions.

    line 81
  9. Distinctive:

    Prefer existing tokens, primitives, icons, and layout patterns when they satisfy the need.

    line 82
  10. Distinctive:

    Add new design variables only when a real requirement is missing.

    line 83
  11. Distinctive:

    Keep spacing, typography, color, elevation, radius, iconography, and motion internally coherent.

    line 84
  12. Distinctive:

    Avoid arbitrary font swaps, default dashboard cards, trendy gradients, decorative glass panels, and borrowed competitor identities as …

    line 85
  13. Distinctive:

    Keep implementation complexity proportional to the design ambition.

    line 86
  14. Distinctive:

    Remove dead styles, placeholder experiments, and unused motion before completion.

    line 87
  15. Distinctive:

    Avoid horizontal overflow, accidental clipping, unstable layout shifts, and fixed elements that cover actionable content.

    line 121
  16. Distinctive:

    Do not claim accessibility from visual inspection alone.

    line 140
  17. Distinctive:

    Use motion to communicate state, hierarchy, continuity, or feedback.

    line 144
  18. Distinctive:

    Treat work as substantial motion/3D when it includes material specialist complexity such as:

    line 146
  19. Distinctive:

    When implementation is in scope, validate the actual rendered result rather than relying only on code inspection.

    line 170
  20. Distinctive:

    Use the strongest evidence available:

    line 172
  21. Distinctive:

    before/after comparison when modifying an existing interface.

    line 178
  22. Very specific:

    If the browser/runtime environment is unavailable, mark that validation NOT_RUN or BLOCKED_EXTERNAL.

    line 181
  23. Distinctive:

    Do not claim that a design is production-ready, responsive, accessible, or performant without naming the evidence that supports that claim.

    line 195

“Common” is measured: kept by at least 6 in 10 of the 181 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

Frontend design

Design and implement production-grade interfaces that are distinctive because they are grounded in the product, the existing visual language, and the user's task — not because they accumulate decoration.

Operational contract

Objective

Turn a real product task and existing visual evidence into a coherent, implementable interface that works across supported responsive and accessibility states. Distinctiveness must come from product-specific hierarchy, interaction, and visual decisions rather than decorative novelty or a generic template.

Expected inputs

Use the target route/surface, product task, repository/component system, existing tokens/assets/identity evidence, content and data states, supported viewport classes, interaction constraints, accessibility requirements, performance/device limits, and explicit authorization for any identity-level change.

Required outputs

Return the design direction, relevant system snapshot, implemented or reviewed UI surface, responsive behavior, accessibility states, motion/3D classification, changed files or components when implementation is in scope, rendered/interaction evidence actually obtained, and any unresolved runtime or content constraints.

Definition of done

The interface is done when the primary task remains clear and usable across the supported states, established identity is preserved or intentionally changed, responsive/accessibility behavior has explicit evidence, motion complexity is proportionate and safely degraded, and no claim of production readiness, responsiveness, accessibility, or performance exceeds the checks actually performed.

Trigger and exclusion boundary

Use this skill when the request asks to design, implement, revise, or review:

  • application or website UI;
  • visual hierarchy and layout;
  • interaction states and component behavior;
  • responsive behavior across meaningful viewport classes;
  • front-end motion or transitions;
  • a visual system that must integrate with an existing repository.

Do not use it for:

  • copy-only requests; use human-copywriting;
  • brand positioning, naming, or unsupported identity invention; use brand-aware-design when brand evidence exists;
  • backend, database, authentication, infrastructure, or data architecture;
  • generic “make it look modern” advice with no target product, implementation, or evidence boundary;
  • substantial architecture review where the core decision is not primarily UI.

If the request mixes design with another domain, split the work and route each bounded part to the appropriate skill.

Non-negotiable design frame

Before writing UI code, establish five things:

  1. Purpose — what the user needs to accomplish and what the interface must make easier.
  2. Existing identity — inspect the target repository's components, tokens, assets, routes, content, and interaction patterns before proposing a new visual language.
  3. Confirmed constraints — distinguish requirements supported by repository or user evidence from reversible creative decisions.
  4. Dominant direction — choose one primary aesthetic stance and, at most, one supporting influence. Do not use a generic SaaS look as a neutral default.
  5. Differentiation anchor — identify one visible, useful detail that makes the interface recognizable without relying on the logo.

Preserve an existing identity unless the user explicitly authorizes a change. Do not replace established typography, color tokens, logo treatment, density, imagery, or interaction conventions merely to make a screen look more novel.

If the current identity is incomplete, mark the gap and make reversible decisions rather than silently inventing brand rules.

Design feasibility and impact

For a material visual direction, use this five-part check before committing:

  • aesthetic impact;
  • context fit;
  • implementation feasibility;
  • performance and accessibility safety;
  • consistency risk.

A strong direction earns its implementation cost. A visually impressive idea that conflicts with the product, weakens accessibility, creates avoidable performance debt, or breaks a coherent existing system should be reduced or rejected.

Do not turn a heuristic score into false certainty. Explain the tradeoff that actually matters.

Identity-aware implementation

Translate the direction into working product code, not a static mockup detached from the repository.

  • Use semantic HTML and the project's established framework/component conventions.
  • Prefer existing tokens, primitives, icons, and layout patterns when they satisfy the need.
  • Add new design variables only when a real requirement is missing.
  • Keep spacing, typography, color, elevation, radius, iconography, and motion internally coherent.
  • Avoid arbitrary font swaps, default dashboard cards, trendy gradients, decorative glass panels, and borrowed competitor identities as substitutes for product-specific thinking.
  • Keep implementation complexity proportional to the design ambition.
  • Remove dead styles, placeholder experiments, and unused motion before completion.
  • Preserve content hierarchy when localizing or when text length changes materially.

Distinctive does not mean noisy. Minimal does not mean generic.

Responsive behavior

Responsive design is not one media query and a narrower desktop layout.

Define how the following change across widths when relevant:

  • navigation;
  • hierarchy;
  • density;
  • grid/list structure;
  • controls and form layout;
  • tables and dense data;
  • media/imagery;
  • interaction model;
  • fixed/sticky elements;
  • typography and line length.

Preserve the primary task on the smallest supported viewport instead of merely shrinking everything.

Verify at least:

  • the smallest supported viewport;
  • one intermediate width;
  • the widest supported layout;
  • long or localized text;
  • missing/slow media where applicable;
  • touch targets and keyboard navigation;
  • zoom/reflow behavior when relevant.

Avoid horizontal overflow, accidental clipping, unstable layout shifts, and fixed elements that cover actionable content.

Accessibility states

Accessibility is part of the design, not a later audit.

Cover applicable states and behaviors:

  • semantic landmarks and heading structure;
  • meaningful labels and accessible names;
  • logical keyboard order;
  • visible focus;
  • sufficient contrast;
  • non-color status cues;
  • usable touch targets;
  • loading, empty, success, error, disabled, validation, and permission states;
  • reduced-motion behavior;
  • meaningful alt text or intentionally empty alt text for decorative imagery.

Do not claim accessibility from visual inspection alone. Name the checks actually performed and their scope.

Motion and substantial 3D

Use motion to communicate state, hierarchy, continuity, or feedback. Prefer a small number of meaningful interactions over constant animation.

Treat work as substantial motion/3D when it includes material specialist complexity such as:

  • continuous render loops or coordinated timelines;
  • scene graphs with camera, lighting, post-processing, physics, or many objects;
  • custom WebGL/shaders/particle systems;
  • complex model viewers or 3D interaction;
  • authored 3D assets whose loading, memory, device support, or fallback materially affects the product.

For substantial motion or 3D, use adaptive-agent-routing when available to decide whether a specialist/stronger worker or reviewer is justified. Route by actual difficulty, performance risk, and blast radius — not by novelty alone.

The specialist brief should cover:

  • why the effect earns its cost;
  • performance/loading budget;
  • device/browser constraints;
  • graceful degradation;
  • reduced-motion or static fallback;
  • accessibility implications;
  • exact acceptance checks.

Simple CSS transitions, one SVG animation, or an ordinary hover effect do not justify heavyweight specialist routing.

Visual evidence and QA

When implementation is in scope, validate the actual rendered result rather than relying only on code inspection.

Use the strongest evidence available:

  • screenshots at representative viewport sizes;
  • browser inspection of the exact route/state;
  • keyboard/touch interaction checks;
  • responsive behavior under real content;
  • before/after comparison when modifying an existing interface;
  • automated browser checks when they meaningfully cover the acceptance criteria.

If the browser/runtime environment is unavailable, mark that validation NOT_RUN or BLOCKED_EXTERNAL; do not upgrade a static code review into visual approval.

Required delivery notes

For a substantive design or implementation, provide a compact record containing:

  1. Design direction — purpose, evidence boundary, confirmed constraints, and reversible creative decisions.
  2. System snapshot — relevant type, color, spacing, elevation, iconography, and motion conventions.
  3. Responsive states — which viewport classes and content stresses were covered.
  4. Accessibility states — what was checked and what was not.
  5. Motion/3D classification — ordinary or substantial; specialist routing if applicable.
  6. Implementation evidence — changed surfaces and validation actually performed.
  7. Differentiation callout — what makes the result recognizable and why it serves the product instead of acting as decoration.

Do not claim that a design is production-ready, responsive, accessible, or performant without naming the evidence that supports that claim.

Related skills

  • brand-aware-design — establishes visual decisions from explicit brand/product evidence.
  • human-copywriting — owns copy and claim discipline.
  • engineering-verification — verifies implementation claims with fresh evidence.
  • adaptive-agent-routing — routes specialist visual/3D work by difficulty, context, and residual risk.

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.