Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Enforce core architectural standards for scalable Laravel applications. Use when structuring controllers, service layers, action classes, Form Requests, or Service Container bindings in Laravel projects.
.claude/skills/hoangnguyen0403-laravel-architecture/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 95% | 21 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -34% | 0% |
See project structure for recommended directory layout.
php artisan make:request StoreUserRequest).handle() method for use case.AppServiceProvider for swappable implementations.See implementation examples for slim controller, action class, and service container binding patterns.
authorize() and rules() methods.$request->validated() in controller for mass assignment.$request->validate().$request->validate(): Favor Form Request classes.When this skill applies, preserve the following domain terminology or equivalent concrete examples in the answer when relevant:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 21,276 | 17,031 | -20% | 1 | 1 | 0% | 2,813 | 2,462 | -12% | 0 | 0 | — |
case-12 | pass→pass | 21,662 | 14,784 | -32% | 1 | 1 | 0% | 3,026 | 1,960 | -35% | 0 | 0 | — |
case-02 | fail→pass | 21,491 | 10,365 | -52% | 1 | 1 | 0% | 2,891 | 2,431 | -16% | 0 | 0 | — |
case-03 | fail→fail | 22,410 | 15,338 | -32% | 1 | 1 | 0% | 3,292 | 2,077 | -37% | 0 | 0 | — |
case-04 | pass→pass | 16,840 | 14,082 | -16% | 1 | 1 | 0% | 1,827 | 1,923 | +5% | 0 | 0 | — |
case-05 | fail→pass | 20,115 | 15,652 | -22% | 1 | 1 | 0% | 2,677 | 2,311 | -14% | 0 | 0 | — |
case-06 | fail→pass | 18,818 | 15,244 | -19% | 1 | 1 | 0% | 2,397 | 2,175 | -9% | 0 | 0 | — |
case-07 | pass→pass | 16,616 | 48,845 | +194% | 1 | 1 | 0% | 1,876 | 1,622 | -14% | 0 | 0 | — |
case-08 | fail→pass | 24,689 | 15,566 | -37% | 1 | 1 | 0% | 3,956 | 2,603 | -34% | 0 | 0 | — |
case-09 | fail→pass | 12,665 | 14,238 | +12% | 1 | 1 | 0% | 2,177 | 2,078 | -5% | 0 | 0 | — |
case-10 | fail→pass | 18,043 | 14,528 | -19% | 1 | 1 | 0% | 2,354 | 1,853 | -21% | 0 | 0 | — |
case-11 | fail→pass | 15,331 | 16,414 | +7% | 1 | 1 | 0% | 2,631 | 2,077 | -21% | 0 | 0 | — |
case-13 | pass→pass | 18,011 | 13,128 | -27% | 1 | 1 | 0% | 2,513 | 1,727 | -31% | 0 | 0 | — |
case-14 | fail→pass | 19,519 | 14,134 | -28% | 1 | 1 | 0% | 2,324 | 1,928 | -17% | 0 | 0 | — |
case-15 | fail→fail | 17,943 | 14,849 | -17% | 1 | 1 | 0% | 3,255 | 2,040 | -37% | 0 | 0 | — |
case-16 | fail→pass | 18,908 | 7,872 | -58% | 1 | 1 | 0% | 2,593 | 1,785 | -31% | 0 | 0 | — |
case-17 | fail→fail | 17,154 | 12,977 | -24% | 1 | 1 | 0% | 2,082 | 1,531 | -26% | 0 | 0 | — |
case-18 | pass→pass | 8,050 | 8,933 | +11% | 1 | 1 | 0% | 1,289 | 1,821 | +41% | 0 | 0 | — |
case-19 | fail→pass | 13,594 | 14,719 | +8% | 1 | 1 | 0% | 2,383 | 2,107 | -12% | 0 | 0 | — |
case-20 | pass→pass | 10,358 | 14,002 | +35% | 1 | 1 | 0% | 1,763 | 1,705 | -3% | 0 | 0 | — |
case-21 | pass→pass | 6,526 | 11,332 | +74% | 1 | 1 | 0% | 1,078 | 1,433 | +33% | 0 | 0 | — |
case-22 | pass→pass | 10,332 | 13,395 | +30% | 1 | 1 | 0% | 1,556 | 1,703 | +9% | 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 +50 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.