$npx -y skills add softspark/ai-toolkit --skill design-engineeringUI craftsmanship: animation rules, easing, micro-interactions, state polish. Triggers: animation, transition, ease-out, motion, micro-interaction, hover, loading state, UI polish.
| 1 | # Design Engineering Skill |
| 2 | |
| 3 | Based on Emil Kowalski's design engineering philosophy — UI polish, component craftsmanship, and the compound value of invisible details. |
| 4 | |
| 5 | ## Core Principles |
| 6 | |
| 7 | - **Taste is trainable.** Develops through studying exceptional work, reverse-engineering animations, and intentional practice. |
| 8 | - **Invisible details create love.** Most UI refinements users never consciously register — but combined they produce something stunning. |
| 9 | - **Beauty differentiates.** When functionality is table stakes, aesthetic excellence becomes genuine leverage. |
| 10 | |
| 11 | ## Anti-Slop Visual Checklist |
| 12 | |
| 13 | Defaults that signal machine-generated UI. Each is a falsifiable thing to avoid: |
| 14 | |
| 15 | - **Avoid** full-bleed saturated gradient backgrounds (purple-to-pink hero washes). A flat surface or a near-flat tonal shift reads as intentional; a loud gradient reads as a template. |
| 16 | - **Avoid** emoji as load-bearing decoration — emoji standing in for icons, bullet markers, or section badges. Use a real icon set or typographic hierarchy instead. |
| 17 | - **Avoid** the rounded-card-with-left-accent-border cliche repeated across every block. If three sections share that exact treatment, vary the layout or drop the accent. |
| 18 | - **Avoid** hand-drawn fake imagery in SVG (synthetic "photos", invented logos, faux screenshots). Use a real asset or an honest labeled placeholder. |
| 19 | - **Avoid** the overused default font stack (Inter/Roboto on system-ui for everything with no scale or weight intent). Pick type with a reason and lift the actual stack from source when one exists. |
| 20 | |
| 21 | ## Minimum-Scale Floors |
| 22 | |
| 23 | Accessibility-grounded hard thresholds. Going below these is a defect, not a style choice: |
| 24 | |
| 25 | | Context | Floor | Basis | |
| 26 | |---|---|---| |
| 27 | | Slide / presentation body text | ~24px | Readable from the back of a room | |
| 28 | | Print body text | ~12pt | Legible at arm's length on paper | |
| 29 | | Mobile touch targets | 44px × 44px | Apple HIG minimum tappable size | |
| 30 | |
| 31 | Treat these as the lower bound, not the target. Captions and footnotes may approach the floor; primary content should sit comfortably above it. |
| 32 | |
| 33 | ## Context-First Discipline |
| 34 | |
| 35 | High-fidelity work MUST be rooted in real context before any pixels are produced. This mirrors the toolkit's verify-don't-recall ethos: |
| 36 | |
| 37 | - **Read the source first.** Inspect the codebase, design tokens, UI kit, and screenshots that already exist before generating anything. |
| 38 | - **Lift exact values.** Copy real hex codes, the spacing scale, the font stack, and radii straight from source. Do NOT reconstruct token values from memory — recalled values drift. |
| 39 | - **Mock from scratch only as a last resort.** Building a screen with no reference is the fallback when no codebase, kit, or screenshot exists, not the default. |
| 40 | |
| 41 | ## Question-Budget Gate |
| 42 | |
| 43 | Calibrate questions to how bounded the ask is, then proceed: |
| 44 | |
| 45 | - **Rich context + bounded ask** → ask nothing, build. Example: "match this card to the existing dashboard" with the repo in hand. |
| 46 | - **Open ask** → ask before building. Example: "prototype my onboarding" needs goals, target audience, and which dimension to diverge on (UX flow vs. visual treatment vs. copy). Resolve those three, then start. |
| 47 | |
| 48 | ## Explore Many Variations |
| 49 | |
| 50 | For exploratory or open work, produce **3+ atomic variations** across distinct axes, never three tweaks of one idea: |
| 51 | |
| 52 | - Vary on different dimensions: layout, color, type treatment, interaction model. |
| 53 | - Deliberately mix safe matches with at least one novel direction — do not converge early. |
| 54 | - Order them basic → advanced so the reviewer can scan the gradient (the "Design It Twice" premise extended past two). |
| 55 | |
| 56 | ## Match Existing Vocabulary |
| 57 | |
| 58 | When editing a live UI, conform to it instead of imposing a new style: |
| 59 | |
| 60 | - Reverse-engineer the palette, interactive states (hover/active/disabled), motion timing, and shadow/card/density treatment. |
| 61 | - Match the copy tone too — terse product UI and chatty marketing copy are different vocabularies. |
| 62 | - A change that introduces a foreign style is a regression even when it looks good in isolation. |
| 63 | |
| 64 | ## No-Filler Content |
| 65 | |
| 66 | - **Every element earns its place.** No dummy stats, decorative sections, or lorem blocks added just to fill space. If a block has no purpose, cut it. |
| 67 | - **Honest placeholder beats a bad fake.** A labeled placeholder ("[product screenshot]") is better than an invented icon or hand-drawn fake image. Do NOT fabricate assets — ask for the real ones. |
| 68 | |
| 69 | ## Animation Decision Framework |
| 70 | |
| 71 | ### Frequency determines approach |
| 72 | |
| 73 | | Usage Pattern | Strategy | |
| 74 | |---|---| |
| 75 | | 100+ daily | No animation | |
| 76 | | Tens daily | Drastically reduce | |
| 77 | | Occasional | Standard animation | |
| 78 | | Rare/first-time | Add delight | |
| 79 | |
| 80 | **Never animate keyboard-initiated actions** — they repeat hundreds of times daily, making |