Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Project Guidelines Skill (Example)
.claude/skills/dokhacgiakhoa-agent-project-guidelines-example/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 54% | 0% |
This is an example of a project-specific skill. Use this as a template for your own projects.
Based on a real production application: Zenith - AI-powered customer discovery platform.
Reference this skill when working on the specific project it's designed for. Project skills contain:
Tech Stack:
Services:
┌─────────────────────────────────────────────────────────────┐
│ Frontend │
│ Next.js 15 + TypeScript + TailwindCSS │
│ Deployed: Vercel / Cloud Run │
└─────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────┐
│ Backend │
│ FastAPI + Python 3.11 + Pydantic │
│ Deployed: Cloud Run │
└─────────────────────────────────────────────────────────────┘
│
┌───────────────┼───────────────┐
▼ ▼ ▼
┌──────────┐ ┌──────────┐ ┌──────────┐
│ Supabase │ │ Agent │ │ Redis │
│ Database │ │ API │ │ Cache │
└──────────┘ └──────────┘ └──────────┘project/
├── frontend/
│ └── src/
│ ├── app/ # Next.js app router pages
│ │ ├── api/ # API routes
│ │ ├── (auth)/ # Auth-protected routes
│ │ └── workspace/ # Main app workspace
│ ├── components/ # React components
│ │ ├── ui/ # Base UI components
│ │ ├── forms/ # Form components
│ │ └── layouts/ # Layout components
│ ├── hooks/ # Custom React hooks
│ ├── lib/ # Utilities
│ ├── types/ # TypeScript definitions
│ └── config/ # Configuration
│
├── backend/
│ ├── routers/ # FastAPI route handlers
│ ├── models.py # Pydantic models
│ ├── main.py # FastAPI app entry
│ ├── auth_system.py # Authentication
│ ├── database.py # Database operations
│ ├── services/ # Business logic
│ └── tests/ # pytest tests
│
├── deploy/ # Deployment configs
├── docs/ # Documentation
└── scripts/ # Utility scripts| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,462 | 5,767 | -54% | 1 | 1 | 0% | 2,063 | 1,741 | -16% | 0 | 0 | — |
case-02 | fail→pass | 14,062 | 4,694 | -67% | 1 | 1 | 0% | 2,565 | 1,876 | -27% | 0 | 0 | — |
case-03 | fail→pass | 15,531 | 7,611 | -51% | 1 | 1 | 0% | 2,581 | 2,282 | -12% | 0 | 0 | — |
case-04 | pass→pass | 12,956 | 3,048 | -76% | 1 | 1 | 0% | 1,890 | 1,262 | -33% | 0 | 0 | — |
case-05 | fail→pass | 4,721 | 1,897 | -60% | 1 | 1 | 0% | 770 | 1,145 | +49% | 0 | 0 | — |
case-06 | fail→pass | 4,214 | 2,126 | -50% | 1 | 1 | 0% | 729 | 1,124 | +54% | 0 | 0 | — |
case-07 | pass→pass | 10,762 | 2,143 | -80% | 1 | 1 | 0% | 1,738 | 1,225 | -30% | 0 | 0 | — |
case-08 | fail→pass | 12,356 | 2,553 | -79% | 1 | 1 | 0% | 2,352 | 1,310 | -44% | 0 | 0 | — |
case-09 | fail→pass | 4,863 | 1,766 | -64% | 1 | 1 | 0% | 907 | 1,185 | +31% | 0 | 0 | — |
case-10 | fail→pass | 7,107 | 2,141 | -70% | 1 | 1 | 0% | 1,445 | 1,180 | -18% | 0 | 0 | — |
case-11 | fail→pass | 12,804 | 2,134 | -83% | 1 | 1 | 0% | 2,053 | 1,311 | -36% | 0 | 0 | — |
case-12 | fail→pass | 14,493 | 1,673 | -88% | 1 | 1 | 0% | 2,308 | 1,180 | -49% | 0 | 0 | — |
case-13 | fail→pass | 3,780 | 1,590 | -58% | 1 | 1 | 0% | 625 | 1,173 | +88% | 0 | 0 | — |
case-14 | pass→pass | 11,869 | 2,216 | -81% | 1 | 1 | 0% | 2,267 | 1,246 | -45% | 0 | 0 | — |
case-15 | fail→pass | 10,045 | 1,882 | -81% | 1 | 1 | 0% | 1,785 | 1,247 | -30% | 0 | 0 | — |
case-16 | fail→pass | 8,405 | 2,821 | -66% | 1 | 1 | 0% | 1,381 | 1,364 | -1% | 0 | 0 | — |
case-17 | fail→pass | 10,792 | 2,249 | -79% | 1 | 1 | 0% | 1,552 | 1,356 | -13% | 0 | 0 | — |
case-18 | fail→pass | 8,617 | 1,449 | -83% | 1 | 1 | 0% | 1,193 | 1,152 | -3% | 0 | 0 | — |
case-19 | fail→pass | 6,875 | 2,252 | -67% | 1 | 1 | 0% | 1,088 | 1,310 | +20% | 0 | 0 | — |
case-20 | pass→pass | 7,822 | 10,834 | +39% | 1 | 1 | 0% | 1,548 | 2,625 | +70% | 0 | 0 | — |
case-21 | pass→pass | 9,065 | 7,399 | -18% | 1 | 1 | 0% | 1,811 | 2,294 | +27% | 0 | 0 | — |
case-22 | pass→pass | 8,549 | 6,859 | -20% | 1 | 1 | 0% | 1,825 | 2,422 | +33% | 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 +73 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.