Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Design interfaces that adapt gracefully to multiple languages, writing directions, and cultural contexts.
.claude/skills/owl-listener-localization-design/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 57% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 21% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 217% | 0% |
You are an expert in designing UI that works across languages, scripts, and cultures without requiring per-locale redesigns.
You apply localization-aware design principles to ensure components, layouts, and content can be adapted to any target locale without breaking — and that cultural differences in color, iconography, and conventions are accounted for.
The most common localization failure. English is compact; most target languages are longer: | Language | Typical expansion vs English | |---|---| | German | +20–35% | | French | +15–25% | | Finnish | +30–40% | | Arabic (translated) | −20–30% (but RTL and different script) | | Japanese/Chinese | Often shorter, but very different typographic rules | Design for text expansion:
Arabic, Hebrew, Persian, and Urdu read right-to-left. The entire layout mirrors:
margin-inline-start instead of margin-left; padding-block-end instead of padding-bottom — these flip automatically with dir="rtl"text-align: start not text-align: leftWhat does not mirror in RTL:
Color meaning varies significantly by culture: | Color | Western association | Example alternative association | |---|---|---| | Red | Danger, error | Luck/prosperity (China), mourning (South Africa) | | White | Clean, minimal | Mourning (many East Asian cultures) | | Green | Success, go | Unfaithfulness (China), danger (some Middle Eastern contexts) |
Intl.DateTimeFormat / equivalent| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 12,138 | 12,127 | -0% | 1 | 1 | 0% | 2,064 | 3,243 | +57% | 0 | 0 | — |
case-02 | pass→pass | 8,520 | 3,498 | -59% | 1 | 1 | 0% | 1,371 | 1,662 | +21% | 0 | 0 | — |
case-03 | fail→pass | 19,618 | 3,027 | -85% | 1 | 1 | 0% | 1,911 | 1,620 | -15% | 0 | 0 | — |
case-04 | pass→pass | 2,669 | 2,694 | +1% | 1 | 1 | 0% | 513 | 1,624 | +217% | 0 | 0 | — |
case-05 | pass→pass | 3,123 | 3,044 | -3% | 1 | 1 | 0% | 572 | 1,694 | +196% | 0 | 0 | — |
case-06 | pass→pass | 2,358 | 2,121 | -10% | 1 | 1 | 0% | 468 | 1,501 | +221% | 0 | 0 | — |
case-07 | pass→pass | 7,451 | 8,844 | +19% | 1 | 1 | 0% | 1,222 | 2,603 | +113% | 0 | 0 | — |
case-08 | pass→pass | 11,102 | 9,521 | -14% | 1 | 1 | 0% | 1,850 | 2,698 | +46% | 0 | 0 | — |
case-09 | pass→pass | 14,036 | 15,547 | +11% | 1 | 1 | 0% | 2,311 | 3,616 | +56% | 0 | 0 | — |
case-10 | fail→pass | 11,106 | 6,222 | -44% | 1 | 1 | 0% | 1,818 | 2,148 | +18% | 0 | 0 | — |
case-11 | pass→pass | 13,570 | 10,625 | -22% | 1 | 1 | 0% | 2,054 | 2,816 | +37% | 0 | 0 | — |
case-12 | pass→pass | 11,810 | 11,667 | -1% | 1 | 1 | 0% | 1,719 | 2,794 | +63% | 0 | 0 | — |
case-13 | pass→pass | 5,273 | 3,112 | -41% | 1 | 1 | 0% | 824 | 1,568 | +90% | 0 | 0 | — |
case-14 | pass→pass | 10,607 | 11,351 | +7% | 1 | 1 | 0% | 1,744 | 2,967 | +70% | 0 | 0 | — |
case-15 | pass→pass | 4,609 | 6,571 | +43% | 1 | 1 | 0% | 957 | 2,384 | +149% | 0 | 0 | — |
case-16 | pass→pass | 10,605 | 6,361 | -40% | 1 | 1 | 0% | 1,882 | 2,359 | +25% | 0 | 0 | — |
case-17 | fail→fail | 5,195 | 6,077 | +17% | 1 | 1 | 0% | 702 | 1,998 | +185% | 0 | 0 | — |
case-18 | pass→pass | 11,916 | 13,574 | +14% | 1 | 1 | 0% | 2,101 | 3,390 | +61% | 0 | 0 | — |
case-19 | pass→pass | 7,815 | 4,274 | -45% | 1 | 1 | 0% | 1,200 | 1,764 | +47% | 0 | 0 | — |
case-20 | pass→pass | 9,213 | 13,893 | +51% | 1 | 1 | 0% | 1,857 | 3,847 | +107% | 0 | 0 | — |
case-21 | pass→pass | 10,459 | 12,169 | +16% | 1 | 1 | 0% | 2,319 | 3,743 | +61% | 0 | 0 | — |
case-22 | pass→pass | 6,684 | 8,882 | +33% | 1 | 1 | 0% | 1,176 | 2,669 | +127% | 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 +9 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.