Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Optimal ViewModel and Wizard creation patterns for Avalonia using Zafiro and ReactiveUI.
.claude/skills/davila7-avalonia-viewmodels-zafiro/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -1% | 0% |
This skill provides a set of best practices and patterns for creating ViewModels, Wizards, and managing navigation in Avalonia applications, leveraging the power of ReactiveUI and the Zafiro toolkit.
ReactiveObject, WhenAnyValue, etc.) to handle state and logic.IEnhancedCommand for better command management, including progress reporting and name/text attributes.SlimWizard and WizardBuilder for a declarative and maintainable approach.[Section] attribute to register and discover UI sections automatically.DataTypeViewLocator and manage dependencies in the CompositionRoot.SlimWizard.For real-world implementations, refer to the Angor project:
CreateProjectFlowV2.cs: Excellent example of complex Wizard building.HomeViewModel.cs: Simple section ViewModel using functional-reactive commands.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 16,893 | 10,587 | -37% | 1 | 1 | 0% | 3,539 | 2,509 | -29% | 0 | 0 | — |
case-04 | fail→pass | 15,897 | 8,913 | -44% | 1 | 1 | 0% | 3,184 | 2,227 | -30% | 0 | 0 | — |
case-13 | pass→pass | 14,544 | 8,417 | -42% | 1 | 1 | 0% | 2,583 | 1,854 | -28% | 0 | 0 | — |
case-14 | fail→pass | 12,666 | 9,746 | -23% | 1 | 1 | 0% | 2,295 | 1,929 | -16% | 0 | 0 | — |
case-19 | pass→pass | 11,164 | 9,641 | -14% | 1 | 1 | 0% | 2,231 | 2,434 | +9% | 0 | 0 | — |
case-02 | fail→pass | 15,193 | 11,652 | -23% | 1 | 1 | 0% | 3,050 | 2,664 | -13% | 0 | 0 | — |
case-01 | fail→pass | 15,350 | 12,637 | -18% | 1 | 1 | 0% | 3,061 | 3,024 | -1% | 0 | 0 | — |
case-05 | fail→pass | 13,575 | 5,314 | -61% | 1 | 1 | 0% | 2,789 | 1,335 | -52% | 0 | 0 | — |
case-06 | fail→pass | 12,440 | 5,336 | -57% | 1 | 1 | 0% | 2,649 | 1,298 | -51% | 0 | 0 | — |
case-07 | fail→fail | 10,473 | 1,697 | -84% | 1 | 1 | 0% | 1,932 | 627 | -68% | 0 | 0 | — |
case-08 | pass→pass | 5,448 | 3,207 | -41% | 1 | 1 | 0% | 1,050 | 929 | -12% | 0 | 0 | — |
case-09 | pass→pass | 6,097 | 3,830 | -37% | 1 | 1 | 0% | 1,059 | 997 | -6% | 0 | 0 | — |
case-10 | pass→pass | 14,661 | 9,956 | -32% | 1 | 1 | 0% | 2,787 | 2,261 | -19% | 0 | 0 | — |
case-11 | pass→pass | 15,873 | 10,017 | -37% | 1 | 1 | 0% | 3,358 | 2,439 | -27% | 0 | 0 | — |
case-12 | fail→pass | 12,166 | 8,642 | -29% | 1 | 1 | 0% | 2,335 | 1,990 | -15% | 0 | 0 | — |
case-15 | pass→pass | 13,948 | 9,412 | -33% | 1 | 1 | 0% | 2,423 | 2,002 | -17% | 0 | 0 | — |
case-16 | pass→pass | 12,816 | 8,811 | -31% | 1 | 1 | 0% | 2,517 | 1,862 | -26% | 0 | 0 | — |
case-17 | pass→pass | 14,625 | 9,653 | -34% | 1 | 1 | 0% | 2,706 | 2,147 | -21% | 0 | 0 | — |
case-18 | pass→pass | 13,692 | 7,448 | -46% | 1 | 1 | 0% | 2,383 | 1,673 | -30% | 0 | 0 | — |
case-20 | pass→pass | 10,732 | 5,768 | -46% | 1 | 1 | 0% | 2,362 | 1,502 | -36% | 0 | 0 | — |
case-21 | pass→pass | 16,780 | 16,699 | -0% | 1 | 1 | 0% | 3,849 | 4,254 | +11% | 0 | 0 | — |
case-22 | pass→pass | 13,331 | 10,159 | -24% | 1 | 1 | 0% | 2,615 | 2,285 | -13% | 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 +36 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.