Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Standardized guidelines for Screaming Architecture where folder and file names immediately reveal the business domain. Use when you want the codebase to communicate its purpose at a glance.
.claude/skills/valec3-frontend-architecture-screaming/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -26% | 0% |
❌ Generic (Tech-First)
src/
├── components/
├── hooks/
├── services/
└── utils/Problem: Doesn't tell you what the app does
✅ Screaming (Domain-First)
src/
├── OrderManagement/
│ ├── OrderApproval/
│ ├── OrderTracking/
│ └── OrderHistory/
├── InventoryControl/
│ ├── StockManagement/
│ └── SupplierOrders/
├── CustomerSupport/
│ ├── TicketSystem/
│ └── LiveChat/
└── _infrastructure/ # Technical plumbing1. Business Concepts First
features/
├── ProductCatalog/
│ ├── ProductSearch/
│ │ ├── SearchBar.tsx
│ │ ├── SearchResults.tsx
│ │ └── useProductSearch.ts
│ ├── ProductDetails/
│ └── ProductReviews/2. Technical Folders are Hidden Prefix with _ to de-emphasize:
_infrastructure/
├── _api/
├── _routing/
└── _state/3. Naming Convention
OrderApprovalWorkflow not Workflow4. Example Structure
src/
├── Appointment Scheduling/
│ ├── BookAppointment/
│ ├── CancelAppointment/
│ └── RescheduleAppointment/
├── Patient Records/
│ ├── MedicalHistory/
│ ├── TestResults/
│ └── Prescriptions/
├── Billing/
│ ├── GenerateInvoice/
│ ├── ProcessPayment/
│ └── InsuranceClaims/
└── _infrastructure/
├── _api/
├── _components/ # Generic UI
└── _utils/| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 13,064 | 12,420 | -5% | 1 | 1 | 0% | 2,617 | 2,213 | -15% | 0 | 0 | — |
case-02 | fail→pass | 15,452 | 8,303 | -46% | 1 | 1 | 0% | 2,833 | 2,142 | -24% | 0 | 0 | — |
case-03 | fail→pass | 14,227 | 8,266 | -42% | 1 | 1 | 0% | 2,931 | 2,150 | -27% | 0 | 0 | — |
case-04 | pass→pass | 18,806 | 15,548 | -17% | 1 | 1 | 0% | 3,575 | 3,748 | +5% | 0 | 0 | — |
case-05 | pass→fail | 18,955 | 13,227 | -30% | 1 | 1 | 0% | 2,752 | 3,139 | +14% | 0 | 0 | — |
case-06 | pass→pass | 12,539 | 8,282 | -34% | 1 | 1 | 0% | 2,372 | 2,114 | -11% | 0 | 0 | — |
case-07 | fail→pass | 15,033 | 7,362 | -51% | 1 | 1 | 0% | 2,822 | 1,971 | -30% | 0 | 0 | — |
case-08 | fail→pass | 16,781 | 8,610 | -49% | 1 | 1 | 0% | 2,939 | 2,164 | -26% | 0 | 0 | — |
case-09 | fail→pass | 19,305 | 8,393 | -57% | 1 | 1 | 0% | 3,433 | 2,237 | -35% | 0 | 0 | — |
case-10 | fail→pass | 16,110 | 8,926 | -45% | 1 | 1 | 0% | 2,762 | 2,234 | -19% | 0 | 0 | — |
case-11 | fail→pass | 14,302 | 7,158 | -50% | 1 | 1 | 0% | 2,393 | 1,861 | -22% | 0 | 0 | — |
case-12 | fail→pass | 11,893 | 6,848 | -42% | 1 | 1 | 0% | 2,216 | 1,903 | -14% | 0 | 0 | — |
case-13 | fail→pass | 13,987 | 7,388 | -47% | 1 | 1 | 0% | 2,605 | 1,938 | -26% | 0 | 0 | — |
case-14 | fail→pass | 12,645 | 7,088 | -44% | 1 | 1 | 0% | 2,396 | 1,852 | -23% | 0 | 0 | — |
case-15 | fail→pass | 12,410 | 6,279 | -49% | 1 | 1 | 0% | 2,332 | 1,674 | -28% | 0 | 0 | — |
case-16 | fail→pass | 14,418 | 6,397 | -56% | 1 | 1 | 0% | 2,517 | 1,729 | -31% | 0 | 0 | — |
case-17 | fail→pass | 15,336 | 9,794 | -36% | 1 | 1 | 0% | 2,717 | 2,306 | -15% | 0 | 0 | — |
case-18 | fail→pass | 12,292 | 7,050 | -43% | 1 | 1 | 0% | 2,396 | 1,801 | -25% | 0 | 0 | — |
case-19 | fail→pass | 15,619 | 6,845 | -56% | 1 | 1 | 0% | 2,965 | 1,845 | -38% | 0 | 0 | — |
case-20 | fail→pass | 12,881 | 6,281 | -51% | 1 | 1 | 0% | 2,325 | 1,700 | -27% | 0 | 0 | — |
case-21 | fail→pass | 16,991 | 10,305 | -39% | 1 | 1 | 0% | 2,625 | 2,120 | -19% | 0 | 0 | — |
case-22 | fail→pass | 22,538 | 7,720 | -66% | 1 | 1 | 0% | 3,403 | 2,079 | -39% | 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 +82 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.