Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Mandatory skills, conventions, and behavioral rules for Avalonia UI development using the Zafiro toolkit.
.claude/skills/davila7-avalonia-zafiro-development/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-22 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -20% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 5% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | -12% | 0% |
This skill defines the mandatory conventions and behavioral rules for developing cross-platform applications with Avalonia UI and the Zafiro toolkit. These rules prioritize maintainability, correctness, and a functional-reactive approach.
Result types and avoidance of exceptions for flow control.RefreshableCollection and Validation.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 22,172 | 14,458 | -35% | 1 | 1 | 0% | 3,645 | 3,030 | -17% | 0 | 0 | — |
case-02 | pass→pass | 17,600 | 11,697 | -34% | 1 | 1 | 0% | 3,051 | 2,433 | -20% | 0 | 0 | — |
case-03 | fail→fail | 16,841 | 18,883 | +12% | 1 | 1 | 0% | 3,211 | 4,050 | +26% | 0 | 0 | — |
case-04 | pass→pass | 11,781 | 9,672 | -18% | 1 | 1 | 0% | 2,171 | 2,286 | +5% | 0 | 0 | — |
case-05 | pass→pass | 8,005 | 4,883 | -39% | 1 | 1 | 0% | 1,400 | 1,233 | -12% | 0 | 0 | — |
case-06 | pass→pass | 11,575 | 9,464 | -18% | 1 | 1 | 0% | 1,990 | 1,928 | -3% | 0 | 0 | — |
case-07 | pass→pass | 17,977 | 8,564 | -52% | 1 | 1 | 0% | 3,166 | 1,758 | -44% | 0 | 0 | — |
case-08 | pass→pass | 14,091 | 11,441 | -19% | 1 | 1 | 0% | 2,449 | 2,290 | -6% | 0 | 0 | — |
case-09 | pass→pass | 15,418 | 7,495 | -51% | 1 | 1 | 0% | 2,508 | 1,535 | -39% | 0 | 0 | — |
case-14 | pass→pass | 17,606 | 14,359 | -18% | 1 | 1 | 0% | 2,945 | 2,969 | +1% | 0 | 0 | — |
case-10 | pass→pass | 13,728 | 11,033 | -20% | 1 | 1 | 0% | 2,234 | 2,047 | -8% | 0 | 0 | — |
case-11 | pass→pass | 15,455 | 8,296 | -46% | 1 | 1 | 0% | 2,607 | 1,701 | -35% | 0 | 0 | — |
case-12 | pass→pass | 12,882 | 7,555 | -41% | 1 | 1 | 0% | 2,358 | 1,473 | -38% | 0 | 0 | — |
case-13 | pass→pass | 15,212 | 12,361 | -19% | 1 | 1 | 0% | 2,668 | 2,516 | -6% | 0 | 0 | — |
case-15 | pass→pass | 9,998 | 7,239 | -28% | 1 | 1 | 0% | 1,878 | 1,517 | -19% | 0 | 0 | — |
case-16 | pass→pass | 12,199 | 4,939 | -60% | 1 | 1 | 0% | 1,778 | 1,099 | -38% | 0 | 0 | — |
case-17 | pass→pass | 17,466 | 12,895 | -26% | 1 | 1 | 0% | 2,960 | 2,381 | -20% | 0 | 0 | — |
case-18 | pass→pass | 21,931 | 19,107 | -13% | 1 | 1 | 0% | 3,991 | 3,751 | -6% | 0 | 0 | — |
case-19 | pass→pass | 16,275 | 10,082 | -38% | 1 | 1 | 0% | 2,520 | 1,991 | -21% | 0 | 0 | — |
case-20 | pass→pass | 4,384 | 3,773 | -14% | 1 | 1 | 0% | 708 | 944 | +33% | 0 | 0 | — |
case-21 | pass→pass | 21,218 | 25,004 | +18% | 1 | 1 | 0% | 3,293 | 3,066 | -7% | 0 | 0 | — |
case-22 | fail→pass | 9,023 | 3,510 | -61% | 1 | 1 | 0% | 1,390 | 879 | -37% | 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 +9 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.