Install any skill in seconds. Free to start, no credit card required.
Get Started Free →CRITICAL: Use for Makepad layout system. Triggers on: makepad layout, makepad width, makepad height, makepad flex, makepad padding, makepad margin, makepad flow, makepad align, Fit, Fill, Size, Walk, "how to center in makepad", makepad 布局, makepad 宽度, makepad 对齐, makepad 居中
.claude/skills/sickn33-makepad-layout/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 10 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 8% | 0% |
> Version: makepad-widgets (dev branch) | Last Updated: 2026-01-19 > > Check for updates: https://crates.io/crates/makepad-widgets
You are an expert at Makepad layout system. Help users by:
Walk, Align, Fit, Fill, padding, spacing, or container flow configuration.Refer to the local files for detailed documentation:
./references/layout-system.md - Complete layout reference./references/core-types.md - Walk, Align, Margin, Padding typesBefore answering questions, Claude MUST:
/sync-crate-skills makepad --force 更新文档"rust<View> { width: Fill height: Fill flow: Down padding: 16.0 spacing: 8.0 <Label> { text: "Item 1" } <Label> { text: "Item 2" } }
rust<View> { width: Fill height: Fill align: { x: 0.5, y: 0.5 } <Label> { text: "Centered" } }
rust<View> { width: Fill height: Fit flow: Right spacing: 10.0 align: { y: 0.5 } // Vertically center items <Button> { text: "Left" } <View> { width: Fill } // Spacer <Button> { text: "Right" } }
rust<View> { width: Fill height: Fill flow: Down // Fixed header <View> { width: Fill height: 60.0 } // Flexible content <View> { width: Fill height: Fill // Takes remaining space } }
| Property | Type | Description | |----------|------|-------------| | width | Size | Width of element | | height | Size | Height of element | | padding | Padding | Inner spacing | | margin | Margin | Outer spacing | | flow | Flow | Child layout direction | | spacing | f64 | Gap between children | | align | Align | Child alignment | | clip_x | bool | Clip horizontal overflow | | clip_y | bool | Clip vertical overflow |
| Value | Description | |-------|-------------| | Fit | Size to fit content | | Fill | Fill available space | | 100.0 | Fixed size in pixels | | Fixed(100.0) | Explicit fixed size |
| Value | Description | |-------|-------------| | Down | Top to bottom (column) | | Right | Left to right (row) | | Overlay | Stack on top |
| Value | Position | |-------|----------| | { x: 0.0, y: 0.0 } | Top-left | | { x: 0.5, y: 0.0 } | Top-center | | { x: 1.0, y: 0.0 } | Top-right | | { x: 0.0, y: 0.5 } | Middle-left | | { x: 0.5, y: 0.5 } | Center | | { x: 1.0, y: 0.5 } | Middle-right | | { x: 0.0, y: 1.0 } | Bottom-left | | { x: 0.5, y: 1.0 } | Bottom-center | | { x: 1.0, y: 1.0 } | Bottom-right |
+---------------------------+
| margin |
| +---------------------+ |
| | padding | |
| | +---------------+ | |
| | | content | | |
| | +---------------+ | |
| +---------------------+ |
+---------------------------+Fill for flexible containers, Fit for content-sized elementsflow: Down for vertical, flow: Right for horizontal<View> { width: Fill } as spacer in row layoutsalign to position children within containerFill takes all available space, Fit shrinks to content| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→fail | 16,249 | 14,172 | -13% | 1 | 1 | 0% | 3,596 | 4,465 | +24% | 0 | 0 | — |
case-01 | pass→pass | 10,049 | 8,884 | -12% | 1 | 1 | 0% | 2,228 | 3,304 | +48% | 0 | 0 | — |
case-03 | fail→pass | 12,046 | 5,437 | -55% | 1 | 1 | 0% | 2,442 | 2,563 | +5% | 0 | 0 | — |
case-04 | fail→pass | 10,419 | 4,150 | -60% | 1 | 1 | 0% | 1,933 | 2,184 | +13% | 0 | 0 | — |
case-05 | fail→pass | 14,040 | 6,552 | -53% | 1 | 1 | 0% | 2,823 | 2,721 | -4% | 0 | 0 | — |
case-06 | fail→pass | 8,568 | 4,295 | -50% | 1 | 1 | 0% | 1,854 | 2,245 | +21% | 0 | 0 | — |
case-07 | fail→pass | 9,201 | 4,665 | -49% | 1 | 1 | 0% | 2,142 | 2,322 | +8% | 0 | 0 | — |
case-08 | fail→pass | 8,809 | 8,649 | -2% | 1 | 1 | 0% | 1,712 | 3,246 | +90% | 0 | 0 | — |
case-09 | fail→pass | 6,662 | 2,267 | -66% | 1 | 1 | 0% | 1,241 | 1,920 | +55% | 0 | 0 | — |
case-10 | fail→pass | 7,382 | 4,353 | -41% | 1 | 1 | 0% | 1,555 | 2,311 | +49% | 0 | 0 | — |
case-11 | fail→fail | 7,696 | 3,253 | -58% | 1 | 1 | 0% | 1,518 | 1,990 | +31% | 0 | 0 | — |
case-12 | fail→pass | 10,127 | 4,099 | -60% | 1 | 1 | 0% | 2,033 | 2,247 | +11% | 0 | 0 | — |
case-13 | fail→pass | 7,630 | 5,587 | -27% | 1 | 1 | 0% | 1,560 | 2,532 | +62% | 0 | 0 | — |
case-14 | pass→pass | 8,067 | 5,269 | -35% | 1 | 1 | 0% | 1,436 | 2,491 | +73% | 0 | 0 | — |
case-15 | fail→pass | 6,484 | 5,798 | -11% | 1 | 1 | 0% | 1,316 | 2,428 | +84% | 0 | 0 | — |
case-16 | pass→pass | 9,832 | 9,532 | -3% | 1 | 1 | 0% | 1,685 | 3,141 | +86% | 0 | 0 | — |
case-17 | pass→pass | 5,672 | 4,959 | -13% | 1 | 1 | 0% | 1,093 | 2,281 | +109% | 0 | 0 | — |
case-18 | fail→pass | 7,931 | 3,331 | -58% | 1 | 1 | 0% | 1,588 | 2,098 | +32% | 0 | 0 | — |
case-19 | pass→pass | 8,003 | 6,864 | -14% | 1 | 1 | 0% | 1,551 | 2,652 | +71% | 0 | 0 | — |
case-20 | pass→pass | 16,323 | 17,765 | +9% | 1 | 1 | 0% | 3,366 | 4,519 | +34% | 0 | 0 | — |
case-21 | pass→pass | 11,593 | 11,749 | +1% | 1 | 1 | 0% | 2,275 | 3,888 | +71% | 0 | 0 | — |
case-22 | pass→pass | 9,170 | 6,929 | -24% | 1 | 1 | 0% | 1,711 | 2,748 | +61% | 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 +55 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.