Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert technical documentation writer who creates clear, structured docs including API references, guides, tutorials, and README files. Use when writing documentation, API docs, or developer guides.
.claude/skills/marine-softdrink524-technical-writer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -5% | 0% |
You are an expert technical writer who creates documentation that developers actually want to read.
markdown# Project Name > One-line description ## 🚀 Quick Start [Get running in 30 seconds] ## 📋 Prerequisites [What you need installed] ## 📦 Installation [Step-by-step install] ## 💡 Usage [Code examples] ## 🔧 Configuration [Environment variables, config files] ## 📖 API Reference [Link to full docs] ## 🤝 Contributing [How to contribute] ## 📄 License [License info]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 11,168 | 10,323 | -8% | 1 | 1 | 0% | 2,433 | 2,453 | +1% | 0 | 0 | — |
case-02 | fail→fail | 10,308 | 7,545 | -27% | 1 | 1 | 0% | 2,407 | 2,051 | -15% | 0 | 0 | — |
case-03 | pass→pass | 15,976 | 12,548 | -21% | 1 | 1 | 0% | 3,335 | 3,163 | -5% | 0 | 0 | — |
case-04 | fail→pass | 15,580 | 15,443 | -1% | 1 | 1 | 0% | 3,330 | 3,621 | +9% | 0 | 0 | — |
case-05 | fail→fail | 14,205 | 11,967 | -16% | 1 | 1 | 0% | 2,895 | 3,015 | +4% | 0 | 0 | — |
case-06 | pass→pass | 29,527 | 19,941 | -32% | 1 | 1 | 0% | 5,049 | 4,342 | -14% | 0 | 0 | — |
case-07 | pass→pass | 24,621 | 19,438 | -21% | 1 | 1 | 0% | 4,197 | 4,128 | -2% | 0 | 0 | — |
case-08 | fail→pass | 14,645 | 10,667 | -27% | 1 | 1 | 0% | 3,079 | 2,390 | -22% | 0 | 0 | — |
case-09 | pass→pass | 18,245 | 13,055 | -28% | 1 | 1 | 0% | 3,907 | 3,175 | -19% | 0 | 0 | — |
case-10 | pass→pass | 15,590 | 12,959 | -17% | 1 | 1 | 0% | 3,153 | 2,986 | -5% | 0 | 0 | — |
case-11 | fail→pass | 12,000 | 10,883 | -9% | 1 | 1 | 0% | 2,729 | 2,755 | +1% | 0 | 0 | — |
case-12 | pass→pass | 14,106 | 10,273 | -27% | 1 | 1 | 0% | 3,001 | 2,606 | -13% | 0 | 0 | — |
case-13 | pass→pass | 10,061 | 9,323 | -7% | 1 | 1 | 0% | 2,372 | 2,580 | +9% | 0 | 0 | — |
case-14 | pass→pass | 10,062 | 9,706 | -4% | 1 | 1 | 0% | 2,151 | 2,426 | +13% | 0 | 0 | — |
case-15 | fail→pass | 17,268 | 15,988 | -7% | 1 | 1 | 0% | 4,274 | 4,055 | -5% | 0 | 0 | — |
case-16 | pass→pass | 20,889 | 16,941 | -19% | 1 | 1 | 0% | 4,115 | 3,779 | -8% | 0 | 0 | — |
case-17 | pass→pass | 11,359 | 10,162 | -11% | 1 | 1 | 0% | 2,567 | 2,703 | +5% | 0 | 0 | — |
case-18 | pass→pass | 11,928 | 10,323 | -13% | 1 | 1 | 0% | 2,463 | 2,407 | -2% | 0 | 0 | — |
case-19 | fail→fail | 14,370 | 10,866 | -24% | 1 | 1 | 0% | 2,699 | 2,417 | -10% | 0 | 0 | — |
case-20 | pass→pass | 4,196 | 8,061 | +92% | 1 | 1 | 0% | 975 | 2,127 | +118% | 0 | 0 | — |
case-21 | pass→pass | 10,840 | 8,574 | -21% | 1 | 1 | 0% | 1,817 | 1,941 | +7% | 0 | 0 | — |
case-22 | pass→pass | 10,937 | 7,409 | -32% | 1 | 1 | 0% | 1,711 | 1,509 | -12% | 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 +18 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.