design-system-guardian

SkillMedia

Detects hardcoded colors, spacing, typography, widths, and other raw values that should use design-system tokens instead. Use when a project has an established design system/token set and you want to catch drift from it.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the design-system-guardian skill

What this skill tells your AI

The instructions your AI receives, as published by codebygarv/ai-skills in skills/ui-ux/design-system-guardian/SKILL.md and read by ahel’s review.

Purpose

Catch places where UI code uses raw/hardcoded values (hex colors, pixel spacing, arbitrary font sizes) instead of the project's established design tokens — the specific class of consistency drift that causes a design system to slowly stop being followed.

When to Use

  • The project has an established design-token system (CSS variables, a theme object, Tailwind config, etc.) and you want to check code against it.
  • Reviewing new UI code for token compliance before merge.
  • Auditing an existing codebase for accumulated token drift.

Requires knowing (or being told, or being able to find) the project's actual token set — this skill checks against an existing system, it doesn't invent one from scratch (for that, see UI Auditor for general consistency without an established token baseline).

What to Analyze

  1. Color — hardcoded hex/rgb/named colors where a color token exists for that exact or a near-identical value.
  2. Spacing — raw pixel/rem margin/padding/gap values that don't align to the token scale (e.g. 13px when the scale is 4/8/12/16/24/32).
  3. Typography — arbitrary font-size/font-weight/line-height combinations instead of defined type-scale tokens or text style components.
  4. Sizing — hardcoded widths/heights/border-radius that duplicate an existing token (e.g. a one-off border-radius: 6px when the system defines radius.md: 8px).
  5. Shadows/elevation — custom box-shadow values where an elevation token exists.
  6. Near-misses — values close to but not equal to a token (15px vs. token 16px) — these are often unintentional and worth flagging even more than obviously-custom values.

Output Format

  • Each finding: the raw value used, its file/line, the matching (or closest) token it should be replaced with, and the token's actual name/variable in this project's system.
  • Group by category (Color, Spacing, Typography, Sizing, Shadow).
  • If a value has no reasonably close token, say so explicitly rather than forcing a mismatched token onto it — that's a signal the token system may need a new token, not that the code is wrong.

Avoid

  • Forcing a token fit when the design genuinely calls for a one-off value (e.g. matching a third-party embed's fixed dimensions) — flag it as an intentional exception rather than a violation.
  • Assuming a token system without confirming what it actually is in this project — ask or look for the theme/token source before flagging anything.
  • Flagging token usage inside the design system's own source files (where raw values are expected, since that's where tokens are defined).

Signals

GitHub stars
25
Forks
1
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
design-system-guardian
Source
github.com/codebygarv/ai-skills