Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Design a spacing and layout token system for a Figma design system. Use when asked to create a spacing system, define layout tokens, set up a grid system, build a spacing scale, or establish layout foundations for a Figma file. Produces a complete spacing scale, grid definition, component spacing conventions, and Figma implementation guide.
.claude/skills/mohitagw15856-figma-spacing-system/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 53% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 76% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 15% | 0% |
Produces a complete spacing and layout token system — the foundation that makes a design system consistent and developer handoff unambiguous.
4px or 8px] with rationale. All values must be multiples.
| Token | Value | Use case | |---|---|---| | spacing.none | 0px | Removing space intentionally | | spacing.xs | 4/8px | Icon padding, tight labels | | spacing.sm | 8/12px | Internal component padding compact | | spacing.md | 12/16px | Internal component padding standard | | spacing.lg | 16/24px | Section padding, card internal | | spacing.xl | 24/32px | Between components | | spacing.2xl | 32/48px | Section separation | | spacing.3xl | 48/64px | Page-level breaks | | spacing.4xl | 64/96px | Hero sections |
Mobile (375px): 4 columns, margin value], gutter value] Tablet (768px): 8 columns, margin value], gutter value] Desktop (1440px): 12 columns, margin value], gutter value], max content width value]
| Context | Token | Example | |---|---|---| | Button horizontal padding | spacing.md | Left/right | | Button vertical padding | spacing.sm | Top/bottom | | Card internal padding | spacing.lg | All sides | | Input padding | spacing.sm vertical, spacing.md horizontal | | | Icon gap from label | spacing.xs | | | Section gap | spacing.xl | |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→fail | 31,331 | 18,775 | -40% | 1 | 1 | 0% | 5,866 | 4,607 | -21% | 0 | 0 | — |
case-03 | fail→fail | 32,862 | 21,173 | -36% | 1 | 1 | 0% | 6,095 | 5,031 | -17% | 0 | 0 | — |
case-01 | fail→fail | 30,395 | 18,950 | -38% | 1 | 1 | 0% | 6,243 | 4,803 | -23% | 0 | 0 | — |
case-04 | fail→pass | 17,252 | 18,216 | +6% | 1 | 1 | 0% | 2,859 | 4,373 | +53% | 0 | 0 | — |
case-05 | fail→fail | 12,244 | 16,702 | +36% | 1 | 1 | 0% | 2,206 | 4,078 | +85% | 0 | 0 | — |
case-06 | fail→pass | 12,166 | 10,517 | -14% | 1 | 1 | 0% | 2,100 | 2,859 | +36% | 0 | 0 | — |
case-07 | fail→fail | 12,357 | 14,277 | +16% | 1 | 1 | 0% | 2,406 | 3,752 | +56% | 0 | 0 | — |
case-08 | fail→fail | 14,140 | 17,112 | +21% | 1 | 1 | 0% | 2,476 | 4,172 | +68% | 0 | 0 | — |
case-09 | fail→pass | 14,580 | 18,499 | +27% | 1 | 1 | 0% | 2,428 | 4,276 | +76% | 0 | 0 | — |
case-10 | fail→pass | 14,051 | 14,633 | +4% | 1 | 1 | 0% | 2,385 | 3,633 | +52% | 0 | 0 | — |
case-11 | fail→pass | 15,148 | 13,301 | -12% | 1 | 1 | 0% | 3,011 | 3,473 | +15% | 0 | 0 | — |
case-12 | pass→pass | 15,012 | 19,361 | +29% | 1 | 1 | 0% | 2,621 | 4,376 | +67% | 0 | 0 | — |
case-13 | fail→fail | 15,054 | 19,630 | +30% | 1 | 1 | 0% | 2,882 | 4,827 | +67% | 0 | 0 | — |
case-18 | fail→fail | 13,927 | 26,778 | +92% | 1 | 1 | 0% | 2,882 | 3,822 | +33% | 0 | 0 | — |
case-14 | pass→pass | 8,231 | 18,527 | +125% | 1 | 1 | 0% | 1,440 | 3,956 | +175% | 0 | 0 | — |
case-15 | fail→pass | 14,241 | 14,893 | +5% | 1 | 1 | 0% | 2,425 | 3,727 | +54% | 0 | 0 | — |
case-16 | pass→pass | 21,855 | 19,819 | -9% | 1 | 1 | 0% | 1,180 | 2,268 | +92% | 0 | 0 | — |
case-17 | pass→pass | 15,519 | 20,026 | +29% | 1 | 1 | 0% | 2,545 | 4,674 | +84% | 0 | 0 | — |
case-19 | pass→pass | 3,840 | 3,712 | -3% | 1 | 1 | 0% | 661 | 1,564 | +137% | 0 | 0 | — |
case-20 | pass→pass | 16,261 | 22,317 | +37% | 1 | 1 | 0% | 3,642 | 5,599 | +54% | 0 | 0 | — |
case-21 | pass→pass | 16,511 | 17,393 | +5% | 1 | 1 | 0% | 3,377 | 4,348 | +29% | 0 | 0 | — |
case-22 | pass→pass | 15,567 | 13,587 | -13% | 1 | 1 | 0% | 2,940 | 3,498 | +19% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +27 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.