Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Implement type-safe Jetpack Navigation Compose routes using Kotlin serialization. Use only when defining typed Compose destinations and arguments; defer XML navigation, deep links, animations, and generic auth or bottom-navigation work.
.claude/skills/hoangnguyen0403-android-navigation-type-safe/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 3% | 0% |
| case-19 | ✓→✓ | ＝ Same ✓ | -29% | 0% |
@Serializable objects/classes instead of String routes..toRoute<T>().navigation extension functions).onNavigateToX).@Serializable data class ProductDetail(val productId: Long) and navigate with the typed destination; retrieve it with toRoute<ProductDetail>().When this skill applies, preserve the following domain terminology or equivalent concrete examples in the answer when relevant:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 11,917 | 7,513 | -37% | 1 | 1 | 0% | 2,457 | 1,893 | -23% | 0 | 0 | — |
case-02 | pass→pass | 8,212 | 6,696 | -18% | 1 | 1 | 0% | 1,548 | 1,588 | +3% | 0 | 0 | — |
case-19 | pass→pass | 6,298 | 2,936 | -53% | 1 | 1 | 0% | 1,130 | 800 | -29% | 0 | 0 | — |
case-03 | pass→pass | 10,055 | 7,976 | -21% | 1 | 1 | 0% | 1,912 | 1,987 | +4% | 0 | 0 | — |
case-04 | pass→pass | 12,189 | 7,992 | -34% | 1 | 1 | 0% | 2,512 | 1,926 | -23% | 0 | 0 | — |
case-05 | pass→pass | 7,419 | 5,866 | -21% | 1 | 1 | 0% | 1,463 | 1,386 | -5% | 0 | 0 | — |
case-06 | pass→pass | 9,626 | 4,604 | -52% | 1 | 1 | 0% | 1,661 | 1,216 | -27% | 0 | 0 | — |
case-07 | pass→pass | 13,740 | 9,209 | -33% | 1 | 1 | 0% | 2,352 | 1,940 | -18% | 0 | 0 | — |
case-08 | pass→pass | 14,695 | 9,615 | -35% | 1 | 1 | 0% | 2,633 | 2,140 | -19% | 0 | 0 | — |
case-09 | pass→pass | 9,808 | 7,788 | -21% | 1 | 1 | 0% | 2,030 | 1,766 | -13% | 0 | 0 | — |
case-10 | pass→pass | 12,731 | 9,255 | -27% | 1 | 1 | 0% | 2,410 | 1,939 | -20% | 0 | 0 | — |
case-11 | pass→pass | 11,799 | 6,736 | -43% | 1 | 1 | 0% | 2,297 | 1,490 | -35% | 0 | 0 | — |
case-12 | pass→pass | 14,622 | 8,118 | -44% | 1 | 1 | 0% | 2,426 | 1,950 | -20% | 0 | 0 | — |
case-13 | fail→pass | 14,253 | 8,944 | -37% | 1 | 1 | 0% | 2,552 | 2,262 | -11% | 0 | 0 | — |
case-14 | pass→pass | 8,078 | 5,754 | -29% | 1 | 1 | 0% | 1,518 | 1,285 | -15% | 0 | 0 | — |
case-15 | fail→pass | 13,903 | 10,951 | -21% | 1 | 1 | 0% | 2,449 | 2,155 | -12% | 0 | 0 | — |
case-16 | pass→pass | 12,360 | 10,066 | -19% | 1 | 1 | 0% | 2,322 | 2,255 | -3% | 0 | 0 | — |
case-17 | pass→pass | 11,615 | 7,854 | -32% | 1 | 1 | 0% | 2,275 | 1,810 | -20% | 0 | 0 | — |
case-18 | pass→pass | 7,146 | 5,137 | -28% | 1 | 1 | 0% | 1,280 | 1,231 | -4% | 0 | 0 | — |
case-20 | pass→pass | 11,750 | 11,861 | +1% | 1 | 1 | 0% | 2,505 | 2,811 | +12% | 0 | 0 | — |
case-21 | pass→pass | 10,219 | 9,978 | -2% | 1 | 1 | 0% | 2,001 | 2,112 | +6% | 0 | 0 | — |
case-22 | pass→pass | 7,252 | 7,549 | +4% | 1 | 1 | 0% | 1,420 | 1,770 | +25% | 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 +14 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.