Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate UX microcopy (button labels, error messages, empty states, toasts) following a casual-but-polite voice and tone
.claude/skills/bilal140202-ss-copy/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 50% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 90% | 0% |
/ss-feedbackContext: $0 Description: $ARGUMENTS
DESIGN-LANGUAGE.md sections on Microcopy Tone Guide and UX WritingFormat: [Action verb] + [Object] (optional)
Good: "Place order", "Get started", "Save changes", "Try again"
Bad: "Submit", "OK", "Click here", "Proceed to next step"Format: [Friendly observation] + [Suggested action]
Good: "No activity yet. Create your first project to get started."
Bad: "No data found."Format: [What happened] + [What to do]
Good: "Couldn't load the data. Please try again."
Bad: "Error 500: Internal Server Error"Format: [Confirmation of what happened]
Good: "Saved!", "Changes applied", "Item deleted · Undo"
Bad: "Operation completed successfully"Label: Noun phrase ("Email address", "Password")
Placeholder: Example or hint ("name@example.com")
Helper: Format guidance ("Must be at least 8 characters")
Error: Specific issue ("This email is already registered")Title: [Question about the action]
Body: [Consequence explanation]
Primary: [Action verb] ("Delete", "Confirm")
Secondary: "Close" (not "Cancel" — avoids confusion)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | pass→pass | 10,671 | 6,993 | -34% | 1 | 1 | 0% | 1,801 | 1,904 | +6% | 0 | 0 | — |
case-03 | fail→pass | 7,204 | 5,358 | -26% | 1 | 1 | 0% | 1,054 | 1,563 | +48% | 0 | 0 | — |
case-13 | fail→pass | 9,645 | 7,400 | -23% | 1 | 1 | 0% | 1,672 | 1,972 | +18% | 0 | 0 | — |
case-18 | fail→pass | 8,249 | 6,235 | -24% | 1 | 1 | 0% | 1,375 | 1,864 | +36% | 0 | 0 | — |
case-01 | fail→fail | 12,609 | 8,275 | -34% | 1 | 1 | 0% | 2,086 | 2,068 | -1% | 0 | 0 | — |
case-04 | fail→pass | 6,731 | 4,766 | -29% | 1 | 1 | 0% | 947 | 1,420 | +50% | 0 | 0 | — |
case-05 | pass→pass | 6,430 | 7,269 | +13% | 1 | 1 | 0% | 1,181 | 1,934 | +64% | 0 | 0 | — |
case-06 | fail→fail | 9,181 | 6,738 | -27% | 1 | 1 | 0% | 1,474 | 1,721 | +17% | 0 | 0 | — |
case-07 | pass→pass | 8,255 | 7,072 | -14% | 1 | 1 | 0% | 1,469 | 1,879 | +28% | 0 | 0 | — |
case-08 | pass→pass | 9,420 | 7,258 | -23% | 1 | 1 | 0% | 1,466 | 1,932 | +32% | 0 | 0 | — |
case-09 | pass→pass | 7,697 | 6,035 | -22% | 1 | 1 | 0% | 1,272 | 1,684 | +32% | 0 | 0 | — |
case-10 | pass→pass | 8,921 | 7,497 | -16% | 1 | 1 | 0% | 1,614 | 1,984 | +23% | 0 | 0 | — |
case-11 | fail→fail | 9,313 | 6,835 | -27% | 1 | 1 | 0% | 1,619 | 1,942 | +20% | 0 | 0 | — |
case-12 | pass→pass | 13,013 | 8,752 | -33% | 1 | 1 | 0% | 2,069 | 2,180 | +5% | 0 | 0 | — |
case-14 | pass→pass | 8,177 | 5,806 | -29% | 1 | 1 | 0% | 1,326 | 1,785 | +35% | 0 | 0 | — |
case-15 | pass→pass | 8,836 | 6,948 | -21% | 1 | 1 | 0% | 1,446 | 1,840 | +27% | 0 | 0 | — |
case-16 | fail→pass | 5,905 | 6,188 | +5% | 1 | 1 | 0% | 936 | 1,782 | +90% | 0 | 0 | — |
case-17 | pass→pass | 11,147 | 5,950 | -47% | 1 | 1 | 0% | 1,962 | 1,679 | -14% | 0 | 0 | — |
case-19 | pass→fail | 7,119 | 5,180 | -27% | 1 | 1 | 0% | 1,103 | 1,625 | +47% | 0 | 0 | — |
case-20 | fail→pass | 20,170 | 4,499 | -78% | 1 | 1 | 0% | 2,705 | 1,425 | -47% | 0 | 0 | — |
case-21 | fail→fail | 18,837 | 15,664 | -17% | 1 | 1 | 0% | 3,123 | 3,116 | -0% | 0 | 0 | — |
case-22 | fail→fail | 6,556 | 5,361 | -18% | 1 | 1 | 0% | 1,194 | 1,700 | +42% | 0 | 0 | — |
case-23 | fail→fail | 20,230 | 22,051 | +9% | 1 | 1 | 0% | 4,454 | 5,477 | +23% | 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 +22 percentage points is the difference between those two pass rates over the 23 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.