Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate a composed UI pattern (card layout, list, form section, grid, etc.) using design system primitives
.claude/skills/bitjaru-ss-pattern/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -46% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 86% | 0% |
When .styleseed/project.json and .styleseed/artifacts/index.json exist, resolve the requested artifact ID first, then read only .styleseed/bundles/<artifact-id>.md and .styleseed/manifests/<artifact-id>.json. Never fall back to the global legacy bundle for a registry project. Legacy projects may use .styleseed/effective-rules.md only when no registry exists.
/ss-component/ss-page/ss-flow/ss-tokensPattern type: $0 Description: $ARGUMENTS
.styleseed/effective-rules.md and .styleseed/manifest.json. Resolve first when stale.Then inspect:
components/ui/ for available primitivescomponents/patterns/ for existing patternsss-pattern-surface only when containment is earned.ss-page-gutter or ss-page-padding instead of hardcoded universal gutters.ss-pattern-inset, ss-pattern-control, and ss-pattern-icon for recipe-aware shape./ss-score and verify that the pattern has a functional relationship to its pageinstead of being an isolated pretty card.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 29,734 | 24,367 | -18% | 1 | 1 | 0% | 6,196 | 5,338 | -14% | 0 | 0 | — |
case-02 | fail→fail | 16,679 | 25,261 | +51% | 1 | 1 | 0% | 3,174 | 6,042 | +90% | 0 | 0 | — |
case-03 | fail→fail | 29,346 | 19,727 | -33% | 1 | 1 | 0% | 6,184 | 4,769 | -23% | 0 | 0 | — |
case-04 | pass→pass | 16,083 | 13,362 | -17% | 1 | 1 | 0% | 3,204 | 3,288 | +3% | 0 | 0 | — |
case-05 | pass→pass | 29,349 | 28,150 | -4% | 1 | 1 | 0% | 6,142 | 6,854 | +12% | 0 | 0 | — |
case-06 | pass→pass | 20,263 | 3,746 | -82% | 1 | 1 | 0% | 4,178 | 1,200 | -71% | 0 | 0 | — |
case-07 | fail→pass | 13,187 | 2,756 | -79% | 1 | 1 | 0% | 2,060 | 1,120 | -46% | 0 | 0 | — |
case-08 | fail→pass | 8,822 | 2,339 | -73% | 1 | 1 | 0% | 1,417 | 1,035 | -27% | 0 | 0 | — |
case-09 | fail→pass | 7,206 | 4,953 | -31% | 1 | 1 | 0% | 994 | 1,421 | +43% | 0 | 0 | — |
case-10 | fail→pass | 12,972 | 4,870 | -62% | 1 | 1 | 0% | 1,786 | 1,468 | -18% | 0 | 0 | — |
case-11 | fail→pass | 18,150 | 22,621 | +25% | 1 | 1 | 0% | 3,036 | 5,652 | +86% | 0 | 0 | — |
case-12 | fail→pass | 12,336 | 8,590 | -30% | 1 | 1 | 0% | 1,863 | 2,297 | +23% | 0 | 0 | — |
case-13 | fail→pass | 9,007 | 3,214 | -64% | 1 | 1 | 0% | 1,399 | 1,187 | -15% | 0 | 0 | — |
case-14 | fail→pass | 23,746 | 20,538 | -14% | 1 | 1 | 0% | 4,809 | 5,669 | +18% | 0 | 0 | — |
case-15 | pass→pass | 8,332 | 8,570 | +3% | 1 | 1 | 0% | 1,317 | 2,181 | +66% | 0 | 0 | — |
case-16 | pass→pass | 9,977 | 3,136 | -69% | 1 | 1 | 0% | 1,438 | 1,136 | -21% | 0 | 0 | — |
case-17 | pass→pass | 21,371 | 12,489 | -42% | 1 | 1 | 0% | 4,430 | 3,316 | -25% | 0 | 0 | — |
case-18 | pass→pass | 8,720 | 14,724 | +69% | 1 | 1 | 0% | 1,297 | 3,878 | +199% | 0 | 0 | — |
case-19 | pass→pass | 10,439 | 2,951 | -72% | 1 | 1 | 0% | 1,733 | 1,131 | -35% | 0 | 0 | — |
case-20 | pass→pass | 15,379 | 8,450 | -45% | 1 | 1 | 0% | 2,361 | 2,124 | -10% | 0 | 0 | — |
case-21 | pass→pass | 7,527 | 3,935 | -48% | 1 | 1 | 0% | 1,030 | 1,282 | +24% | 0 | 0 | — |
case-22 | fail→pass | 13,384 | 2,907 | -78% | 1 | 1 | 0% | 2,186 | 1,115 | -49% | 0 | 0 | — |
case-23 | fail→pass | 17,124 | 7,798 | -54% | 1 | 1 | 0% | 2,747 | 2,068 | -25% | 0 | 0 | — |
case-24 | pass→pass | 20,346 | 8,063 | -60% | 1 | 1 | 0% | 2,936 | 2,009 | -32% | 0 | 0 | — |
case-25 | fail→pass | 13,748 | 1,929 | -86% | 1 | 1 | 0% | 1,905 | 927 | -51% | 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. 25 cases were attempted. The headline lift of +44 percentage points is the difference between those two pass rates over the 25 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/17/2026 | +18% |
Other measured skills in the registry, with their headline benchmark lift.