StyleSeed Design Review

← Back to skills

A UI reads as "AI-generated" not because the components are ugly, but because the **parts don't agree with each other** — mixed corner radii, three accent colors, pure-black text, no hierarchy, missing states, robotic copy. This skill reviews a UI file (or a whole directory) against a concrete design rubric, scores it 0–100, and returns a prioritized fix list. It reviews and recommends; it never edits or deletes without you asking.

Category: Frontend & UI/UX
Repo: antigravity-awesome-skills
Path: skills/styleseed-design-review/SKILL.md
Updated: 7/5/2026, 4:58:46 PM

AI Summary

A UI reads as "AI-generated" not because the components are ugly, but because the **parts don't agree with each other** — mixed corner radii, three accent colors, pure-black text, no hierarchy, missing states, robotic copy. This skill reviews a UI file (or a whole directory) against a concrete design rubric, scores it 0–100, and returns a prioritized fix list. It reviews and recommends; it never edits or deletes without you asking. It is useful for React and Next.js, CSS and design systems, UI components, accessibility, and frontend polish. Source: antigravity-awesome-skills (skills/styleseed-design-review/SKILL.md).

StyleSeed Design Review

Overview

A UI reads as "AI-generated" not because the components are ugly, but because the parts don't agree with each other — mixed corner radii, three accent colors, pure-black text, no hierarchy, missing states, robotic copy. This skill reviews a UI file (or a whole directory) against a concrete design rubric, scores it 0–100, and returns a prioritized fix list. It reviews and recommends; it never edits or deletes without you asking.

Full rule set (74 rules) and components: https://github.com/bitjaru/styleseed

When to use

  • A React / Tailwind / HTML UI "looks off," generic, or unfinished and you can't say why.
  • You want a design score / pre-ship check.
  • The user asks to make UI "look professional / polished / designed, not AI-generated."
  • After generating UI, to verify it before shipping.

How to review

Read the file(s). Score these seven categories (total 100); start each at full marks and subtract for violations you can cite by line. Be specific and evidence-based.

1. Coherence — 20 (the #1 "AI-generated" tell)

One choice per axis, applied everywhere. Deduct for each mixed axis:

  • mixed corner radii — e.g. a sharp card with pill buttons (−6)
  • two or more accent colors used for emphasis (−5)
  • emoji used as UI icons (🚗🧺⭐ as list/nav/status/category markers) — injects many uncontrolled hues; use one line-icon set in currentColor (−6)
  • mixed shadow languages / light directions (−3)
  • mixed icon families, fill modes, or stroke weights (−3)
  • inconsistent control heights (buttons/inputs differ) (−3)

2. Color discipline — 16

  • pure black (#000 / text-black) text — the refined black is ~#2A2A2A (−4 each, cap −8)
  • hardcoded hex where a semantic token exists (−2 each, cap −6)
  • a normal / OK / default ("보통") state shown in a status color instead of neutral grey (−4)
  • status color on most/every row (no severity hierarchy — color should mark the minority that needs attention) (−4)
  • decorative hues — gold stars, rainbow category dots, a different color per card — instead of accent/grey (−3)
  • status conveyed by color alone, no icon/text (−4)
  • contrast below WCAG AA (4.5:1 body, 3:1 large/UI) (−6)

3. Hierarchy & typography — 16

  • number and its unit not ~2:1 (48px number / 24px unit) (−4)
  • everything the same size and weight, no clear primary (−5)
  • arbitrary font sizes; no scale (−4)
  • wrong line-height (loose on display, cramped on body) (−3)

4. Layout & spacing — 12

  • content on a bare page background, not in cards (−6)
  • off-grid spacing (7/13/19px instead of an 8px scale) (−3)
  • the gap around a group not larger than the gap inside it (−3)
  • the same section type repeated in a row (−4)

5. States — 12

  • missing empty / loading / error state on a data surface (−5 each, cap −10)
  • empty state with no next action; error that blames instead of helping (−4)

6. UX writing — 12

  • buttons that don't name the action ("Submit" / "OK" instead of "Send $2,400") (−4)
  • error copy that blames or uses system-speak ("Invalid input", "An error occurred") (−4)
  • two terms for one concept (delete vs remove); filler words ("please", "successfully") (−2)

7. Motion & polish — 12

  • ad-hoc fades instead of one consistent, named feel (−3)
  • motion that delays content or blocks an action (−4)
  • no prefers-reduced-motion handling on custom motion (−3)
  • a single hard black shadow instead of a layered, low-opacity, tinted one (−2)

Clamp each category at 0; sum to a total. Bands: 90+ A · 80–89 B · 70–79 C · 60–69 D · <60 F.

Output format

## Design Score: 72 / 100   (src/Dashboard.tsx)   C

Coherence            13/20   sharp cards (l.22) + pill buttons (l.48); 3 accent hues
Color discipline     12/16   #000 headings (l.12, 40)
Hierarchy & type     15/16   number/unit 1:1 on hero (l.18)
Layout & spacing     10/12   two identical KPI rows (l.22-31)
States                7/12   no empty/loading state on the orders list
UX writing            8/12   "Submit" button (l.55); "Invalid input" (l.61)
Motion & polish      10/12   one hard black shadow (l.22)

### Fix first (highest score gain)
1. Unify radius (pick soft 8–12px) + collapse to one accent   → +11 coherence/color
2. Add empty + loading states to the orders list              → +7  states
3. Rename "Submit" → "Send $2,400"; "Invalid input" → "Check the card number" → +6 copy

Re-score after: ~90 / 100.

Rules

  • Review from real evidence (cite line numbers); never guess.
  • Order the fix list by score gain, not severity alone — fastest path to a better number.
  • For a directory: one-line score per file, then the lowest file's full breakdown.
  • Don't auto-edit. This skill measures and recommends. Apply fixes only when asked.
  • Use it as a quality gate: review right after generating UI, apply the fix list, and re-review until the score clears ~80 before showing the user — no first-draft, incoherent UI (rainbow status lists, emoji icons, two accents, missing states) should reach them. The bar is a floor, not a ceiling: clear 80 and ship; don't chase 100 to delay.

Based on StyleSeed — an open-source (MIT) design engine that gives Claude Code, Cursor, and Codex design judgment so AI-built UI stops looking generated. Full 74-rule reference, components, brand skins, and motion: https://github.com/bitjaru/styleseed

Limitations

  • Use this skill only when the task clearly matches its upstream source and local project context.
  • Verify commands, generated code, dependencies, credentials, and external service behavior before applying changes.
  • Do not treat examples as a substitute for environment-specific tests, security review, or user approval for destructive or costly actions.

Related skills