Install any skill in seconds. Free to start, no credit card required.
Get Started Free →expo-ui-jetpack-compose
.claude/skills/expo-ui-jetpack-compose/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -50% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -36% | 0% |
> The instructions in this skill apply to SDK 55 only. For other SDK versions, refer to the Expo UI Jetpack Compose docs for that version for the most accurate information.
@expo/ui/jetpack-compose.Host, or translating Jetpack Compose patterns into Expo UI code.bashnpx expo install @expo/ui
A native rebuild is required after installation (npx expo run:android).
@expo/ui/jetpack-compose, modifiers from @expo/ui/jetpack-compose/modifiers.Host. Use <Host matchContents> for intrinsic sizing, or <Host style={{ flex: 1 }}> when you need explicit size (e.g. as a parent of LazyColumn). Example:jsximport { Host, Column, Button, Text } from "@expo/ui/jetpack-compose"; import { fillMaxWidth, paddingAll } from "@expo/ui/jetpack-compose/modifiers"; <Host matchContents> <Column verticalArrangement={{ spacedBy: 8 }} modifiers={[fillMaxWidth(), paddingAll(16)]}> <Text style={{ typography: "titleLarge" }}>Hello</Text> <Button onPress={() => alert("Pressed!")}>Press me</Button> </Column> </Host>;
ScrollView/FlatList for scrollable lists. Wrap in <Host style={{ flex: 1 }}>.<Icon source={require('./icon.xml')} size={24} /> with Android XML vector drawables from Material Symbols.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 15,814 | 4,364 | -72% | 1 | 1 | 0% | 3,002 | 1,506 | -50% | 0 | 0 | — |
case-02 | fail→pass | 13,323 | 12,062 | -9% | 1 | 1 | 0% | 2,919 | 2,725 | -7% | 0 | 0 | — |
case-03 | fail→pass | 11,571 | 5,514 | -52% | 1 | 1 | 0% | 2,129 | 1,809 | -15% | 0 | 0 | — |
case-04 | fail→fail | 8,325 | 2,916 | -65% | 1 | 1 | 0% | 1,473 | 1,142 | -22% | 0 | 0 | — |
case-05 | fail→pass | 8,407 | 6,808 | -19% | 1 | 1 | 0% | 1,429 | 1,259 | -12% | 0 | 0 | — |
case-06 | fail→fail | 9,583 | 4,325 | -55% | 1 | 1 | 0% | 1,665 | 1,227 | -26% | 0 | 0 | — |
case-07 | fail→fail | 10,416 | 3,484 | -67% | 1 | 1 | 0% | 1,821 | 1,332 | -27% | 0 | 0 | — |
case-08 | fail→pass | 13,414 | 4,745 | -65% | 1 | 1 | 0% | 2,530 | 1,626 | -36% | 0 | 0 | — |
case-09 | fail→pass | 7,810 | 2,774 | -64% | 1 | 1 | 0% | 1,369 | 1,188 | -13% | 0 | 0 | — |
case-10 | fail→fail | 5,762 | 2,769 | -52% | 1 | 1 | 0% | 1,063 | 1,136 | +7% | 0 | 0 | — |
case-11 | fail→fail | 12,190 | 4,708 | -61% | 1 | 1 | 0% | 2,159 | 1,558 | -28% | 0 | 0 | — |
case-12 | fail→pass | 11,167 | 2,948 | -74% | 1 | 1 | 0% | 2,221 | 1,207 | -46% | 0 | 0 | — |
case-13 | fail→pass | 14,098 | 5,838 | -59% | 1 | 1 | 0% | 2,384 | 1,848 | -22% | 0 | 0 | — |
case-14 | pass→pass | 9,301 | 3,295 | -65% | 1 | 1 | 0% | 1,661 | 1,222 | -26% | 0 | 0 | — |
case-15 | fail→fail | 9,067 | 2,964 | -67% | 1 | 1 | 0% | 1,496 | 1,024 | -32% | 0 | 0 | — |
case-16 | fail→pass | 11,855 | 5,773 | -51% | 1 | 1 | 0% | 2,198 | 1,804 | -18% | 0 | 0 | — |
case-17 | fail→pass | 7,767 | 2,985 | -62% | 1 | 1 | 0% | 1,423 | 1,157 | -19% | 0 | 0 | — |
case-18 | fail→pass | 13,088 | 5,100 | -61% | 1 | 1 | 0% | 2,664 | 1,658 | -38% | 0 | 0 | — |
case-19 | fail→pass | 10,772 | 3,312 | -69% | 1 | 1 | 0% | 1,978 | 1,246 | -37% | 0 | 0 | — |
case-20 | fail→pass | 7,832 | 2,623 | -67% | 1 | 1 | 0% | 1,199 | 1,049 | -13% | 0 | 0 | — |
case-21 | fail→fail | 6,520 | 3,068 | -53% | 1 | 1 | 0% | 1,227 | 1,164 | -5% | 0 | 0 | — |
case-22 | fail→pass | 12,082 | 4,462 | -63% | 1 | 1 | 0% | 2,287 | 1,457 | -36% | 0 | 0 | — |
case-23 | fail→fail | 14,553 | 8,355 | -43% | 1 | 1 | 0% | 3,005 | 2,376 | -21% | 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. 23 cases were attempted. The headline lift of +61 percentage points is the difference between those two pass rates over the 23 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.