Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Implement Jetpack Navigation Component with XML graphs and SafeArgs for type-safe fragment navigation. Use when working with XML-based navigation or SafeArgs in legacy Android projects; defer drawer and other UI-chrome work to UI skills.
.claude/skills/hoangnguyen0403-android-legacy-navigation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-22 | ✗→✓ | ▲ Improved | -13% | 0% |
NavHostFragment.See XML graph & SafeArgs examples for NavHostFragment setup.
navigation/ resources (e.g., nav_auth.xml, nav_main.xml) to keep graphs readable.<deepLink> in graph, not AndroidManifest intent filters.See XML graph & SafeArgs examples for type-safe navigation usage.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | pass→pass | 11,304 | 8,196 | -27% | 1 | 1 | 0% | 2,151 | 1,948 | -9% | 0 | 0 | — |
case-01 | fail→pass | 14,929 | 10,248 | -31% | 1 | 1 | 0% | 2,960 | 2,412 | -19% | 0 | 0 | — |
case-02 | pass→pass | 14,374 | 8,248 | -43% | 1 | 1 | 0% | 2,261 | 2,005 | -11% | 0 | 0 | — |
case-03 | pass→pass | 7,647 | 4,163 | -46% | 1 | 1 | 0% | 1,424 | 984 | -31% | 0 | 0 | — |
case-04 | fail→pass | 9,857 | 6,149 | -38% | 1 | 1 | 0% | 1,921 | 1,522 | -21% | 0 | 0 | — |
case-05 | pass→pass | 11,415 | 4,489 | -61% | 1 | 1 | 0% | 2,276 | 1,100 | -52% | 0 | 0 | — |
case-06 | pass→pass | 14,124 | 8,122 | -42% | 1 | 1 | 0% | 2,555 | 1,817 | -29% | 0 | 0 | — |
case-07 | pass→pass | 9,808 | 9,697 | -1% | 1 | 1 | 0% | 1,885 | 1,934 | +3% | 0 | 0 | — |
case-09 | pass→pass | 3,317 | 2,141 | -35% | 1 | 1 | 0% | 600 | 567 | -6% | 0 | 0 | — |
case-10 | pass→pass | 3,674 | 3,125 | -15% | 1 | 1 | 0% | 677 | 758 | +12% | 0 | 0 | — |
case-11 | pass→pass | 3,423 | 2,587 | -24% | 1 | 1 | 0% | 615 | 662 | +8% | 0 | 0 | — |
case-12 | pass→pass | 10,328 | 4,075 | -61% | 1 | 1 | 0% | 1,854 | 863 | -53% | 0 | 0 | — |
case-13 | fail→pass | 13,485 | 7,833 | -42% | 1 | 1 | 0% | 2,484 | 1,854 | -25% | 0 | 0 | — |
case-14 | pass→pass | 7,174 | 3,260 | -55% | 1 | 1 | 0% | 1,362 | 840 | -38% | 0 | 0 | — |
case-15 | pass→pass | 14,014 | 7,955 | -43% | 1 | 1 | 0% | 2,615 | 1,808 | -31% | 0 | 0 | — |
case-16 | pass→pass | 8,301 | 5,603 | -33% | 1 | 1 | 0% | 1,284 | 1,336 | +4% | 0 | 0 | — |
case-17 | pass→pass | 13,952 | 6,112 | -56% | 1 | 1 | 0% | 2,113 | 1,361 | -36% | 0 | 0 | — |
case-18 | pass→pass | 7,603 | 3,970 | -48% | 1 | 1 | 0% | 1,372 | 806 | -41% | 0 | 0 | — |
case-19 | pass→pass | 2,821 | 2,730 | -3% | 1 | 1 | 0% | 532 | 637 | +20% | 0 | 0 | — |
case-20 | fail→pass | 15,030 | 3,780 | -75% | 1 | 1 | 0% | 2,282 | 857 | -62% | 0 | 0 | — |
case-21 | pass→pass | 3,351 | 2,660 | -21% | 1 | 1 | 0% | 584 | 664 | +14% | 0 | 0 | — |
case-22 | fail→pass | 13,876 | 8,915 | -36% | 1 | 1 | 0% | 2,353 | 2,048 | -13% | 0 | 0 | — |
case-23 | pass→pass | 14,115 | 10,566 | -25% | 1 | 1 | 0% | 2,672 | 2,333 | -13% | 0 | 0 | — |
case-24 | pass→pass | 12,453 | 9,341 | -25% | 1 | 1 | 0% | 2,277 | 1,973 | -13% | 0 | 0 | — |
case-25 | pass→pass | 12,788 | 10,331 | -19% | 1 | 1 | 0% | 2,649 | 2,176 | -18% | 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 +20 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.
Other measured skills in the registry, with their headline benchmark lift.