Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use this skill when writing, rewriting, or improving marketing copy for any page (homepage, landing page, pricing, feature, product, or about page). This skill produces clear, compelling, and testable copy while enforcing alignment, honesty, and conversion best practices.
.claude/skills/dokhacgiakhoa-copywriting/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 106% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 19% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 57% | 0% |
Produce clear, credible, and action-oriented marketing copy that aligns with user intent and business goals.
This skill exists to prevent:
You may not fabricate claims, statistics, testimonials, or guarantees.
You are operating as an expert conversion copywriter, not a brand poet.
Your job is to help the right reader take the right action.
Before writing any copy, gather or confirm the following. If information is missing, ask for it before proceeding.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 15,165 | 7,815 | -48% | 1 | 1 | 0% | 2,350 | 1,855 | -21% | 0 | 0 | — |
case-01 | fail→pass | 12,839 | 38,198 | +198% | 1 | 1 | 0% | 2,131 | 1,830 | -14% | 0 | 0 | — |
case-02 | fail→fail | 12,462 | 21,702 | +74% | 1 | 1 | 0% | 2,160 | 4,035 | +87% | 0 | 0 | — |
case-04 | fail→fail | 9,121 | 13,760 | +51% | 1 | 1 | 0% | 1,437 | 2,781 | +94% | 0 | 0 | — |
case-05 | fail→fail | 7,793 | 9,946 | +28% | 1 | 1 | 0% | 1,159 | 2,213 | +91% | 0 | 0 | — |
case-06 | fail→pass | 9,301 | 15,825 | +70% | 1 | 1 | 0% | 1,513 | 3,117 | +106% | 0 | 0 | — |
case-07 | fail→pass | 12,712 | 12,691 | -0% | 1 | 1 | 0% | 2,016 | 2,392 | +19% | 0 | 0 | — |
case-08 | fail→pass | 11,382 | 11,977 | +5% | 1 | 1 | 0% | 1,728 | 2,705 | +57% | 0 | 0 | — |
case-09 | fail→fail | 14,480 | 10,120 | -30% | 1 | 1 | 0% | 1,967 | 2,133 | +8% | 0 | 0 | — |
case-10 | fail→fail | 4,938 | 8,792 | +78% | 1 | 1 | 0% | 796 | 1,824 | +129% | 0 | 0 | — |
case-11 | fail→fail | 7,598 | 17,985 | +137% | 1 | 1 | 0% | 1,194 | 3,531 | +196% | 0 | 0 | — |
case-12 | fail→fail | 6,989 | 13,035 | +87% | 1 | 1 | 0% | 1,150 | 2,623 | +128% | 0 | 0 | — |
case-13 | fail→pass | 13,721 | 14,841 | +8% | 1 | 1 | 0% | 1,963 | 2,318 | +18% | 0 | 0 | — |
case-14 | pass→pass | 14,886 | 11,717 | -21% | 1 | 1 | 0% | 2,418 | 2,372 | -2% | 0 | 0 | — |
case-15 | pass→pass | 17,338 | 19,092 | +10% | 1 | 1 | 0% | 3,126 | 3,368 | +8% | 0 | 0 | — |
case-16 | pass→pass | 8,759 | 6,624 | -24% | 1 | 1 | 0% | 1,474 | 1,803 | +22% | 0 | 0 | — |
case-17 | pass→pass | 14,957 | 14,853 | -1% | 1 | 1 | 0% | 2,587 | 2,745 | +6% | 0 | 0 | — |
case-18 | pass→pass | 8,707 | 8,231 | -5% | 1 | 1 | 0% | 1,506 | 1,917 | +27% | 0 | 0 | — |
case-19 | pass→pass | 11,065 | 10,287 | -7% | 1 | 1 | 0% | 1,666 | 1,940 | +16% | 0 | 0 | — |
case-20 | fail→fail | 4,496 | 7,474 | +66% | 1 | 1 | 0% | 772 | 1,679 | +117% | 0 | 0 | — |
case-21 | fail→pass | 9,908 | 6,285 | -37% | 1 | 1 | 0% | 1,440 | 1,437 | -0% | 0 | 0 | — |
case-22 | fail→fail | 10,517 | 15,602 | +48% | 1 | 1 | 0% | 1,650 | 2,630 | +59% | 0 | 0 | — |
case-23 | fail→pass | 5,881 | 11,686 | +99% | 1 | 1 | 0% | 941 | 2,227 | +137% | 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. 23 cases were attempted. The headline lift of +35 percentage points is the difference between those two pass rates over the 23 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.