Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Master Flutter development with Dart 3, advanced widgets, and multi-platform deployment. Handles state management, animations, testing, and performance optimization for mobile, web, desktop, and embedded platforms. Use PROACTIVELY for Flutter architecture, UI implementation, or cross-platform features.
.claude/skills/dokhacgiakhoa-flutter-expert/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 14% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | 7% | 0% |
resources/implementation-playbook.md.You are a Flutter expert specializing in high-performance, multi-platform applications with deep knowledge of the Flutter 2025 ecosystem.
Expert Flutter developer specializing in Flutter 3.x+, Dart 3.x, and comprehensive multi-platform development. Masters advanced widget composition, performance optimization, and platform-specific integrations while maintaining a unified codebase across mobile, web, desktop, and embedded platforms.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 28,163 | 19,212 | -32% | 1 | 1 | 0% | 5,265 | 3,483 | -34% | 0 | 0 | — |
case-02 | fail→fail | 43,531 | 24,705 | -43% | 1 | 1 | 0% | 7,452 | 5,066 | -32% | 0 | 0 | — |
case-03 | fail→fail | 23,900 | 27,248 | +14% | 1 | 1 | 0% | 4,353 | 4,754 | +9% | 0 | 0 | — |
case-04 | fail→fail | 24,588 | 18,693 | -24% | 1 | 1 | 0% | 3,847 | 4,502 | +17% | 0 | 0 | — |
case-05 | fail→pass | 14,466 | 16,821 | +16% | 1 | 1 | 0% | 2,689 | 3,559 | +32% | 0 | 0 | — |
case-06 | pass→pass | 17,833 | 17,647 | -1% | 1 | 1 | 0% | 2,865 | 3,254 | +14% | 0 | 0 | — |
case-07 | pass→pass | 19,635 | 14,714 | -25% | 1 | 1 | 0% | 2,851 | 3,053 | +7% | 0 | 0 | — |
case-08 | pass→pass | 11,959 | 8,747 | -27% | 1 | 1 | 0% | 1,674 | 1,887 | +13% | 0 | 0 | — |
case-18 | pass→pass | 19,461 | 21,252 | +9% | 1 | 1 | 0% | 3,511 | 4,576 | +30% | 0 | 0 | — |
case-09 | pass→pass | 19,255 | 19,928 | +3% | 1 | 1 | 0% | 3,057 | 3,735 | +22% | 0 | 0 | — |
case-10 | pass→pass | 7,320 | 9,266 | +27% | 1 | 1 | 0% | 1,148 | 1,554 | +35% | 0 | 0 | — |
case-11 | fail→pass | 11,258 | 6,760 | -40% | 1 | 1 | 0% | 1,537 | 1,601 | +4% | 0 | 0 | — |
case-12 | pass→pass | 21,488 | 23,923 | +11% | 1 | 1 | 0% | 4,190 | 4,363 | +4% | 0 | 0 | — |
case-13 | pass→pass | 22,060 | 17,164 | -22% | 1 | 1 | 0% | 2,668 | 3,797 | +42% | 0 | 0 | — |
case-14 | pass→pass | 16,354 | 21,973 | +34% | 1 | 1 | 0% | 2,557 | 2,893 | +13% | 0 | 0 | — |
case-15 | pass→pass | 15,317 | 17,113 | +12% | 1 | 1 | 0% | 2,624 | 2,903 | +11% | 0 | 0 | — |
case-16 | pass→pass | 15,840 | 14,714 | -7% | 1 | 1 | 0% | 2,669 | 2,900 | +9% | 0 | 0 | — |
case-17 | pass→pass | 14,203 | 12,659 | -11% | 1 | 1 | 0% | 2,214 | 2,199 | -1% | 0 | 0 | — |
case-19 | fail→pass | 12,941 | 10,915 | -16% | 1 | 1 | 0% | 2,180 | 2,266 | +4% | 0 | 0 | — |
case-20 | pass→pass | 13,834 | 19,454 | +41% | 1 | 1 | 0% | 2,513 | 2,589 | +3% | 0 | 0 | — |
case-21 | pass→pass | 16,041 | 12,726 | -21% | 1 | 1 | 0% | 2,393 | 2,466 | +3% | 0 | 0 | — |
case-22 | pass→pass | 17,382 | 13,007 | -25% | 1 | 1 | 0% | 2,963 | 3,127 | +6% | 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 +14 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.