Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate data-driven user personas for UX research and product design. Usage: /persona generate [options]
.claude/skills/alirezarezvani-persona/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -76% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -66% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 5% | 0% |
Generate structured user personas with demographics, goals, pain points, and behavioral patterns.
/persona generate Generate persona (interactive)
/persona generate json Generate persona as JSONInteractive mode prompts for product context. Alternatively, provide context inline:
/persona generate
> Product: B2B project management tool
> Target: Engineering managers at mid-size companies
> Key problem: Cross-team visibility/persona generate
/persona generate json
/persona generate json > persona-eng-manager.jsonproduct-team/skills/ux-researcher-designer/scripts/persona_generator.py — Persona generator (positional json arg for JSON output)> product-team/skills/ux-researcher-designer/SKILL.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | fail→pass | 8,646 | 4,584 | -47% | 1 | 1 | 0% | 1,444 | 972 | -33% | 0 | 0 | — |
case-01 | fail→pass | 8,498 | 2,288 | -73% | 1 | 1 | 0% | 1,607 | 697 | -57% | 0 | 0 | — |
case-02 | fail→pass | 8,535 | 1,014 | -88% | 1 | 1 | 0% | 1,425 | 348 | -76% | 0 | 0 | — |
case-03 | fail→pass | 7,402 | 1,446 | -80% | 1 | 1 | 0% | 1,400 | 471 | -66% | 0 | 0 | — |
case-04 | pass→pass | 3,478 | 3,538 | +2% | 1 | 1 | 0% | 684 | 779 | +14% | 0 | 0 | — |
case-05 | fail→pass | 4,234 | 3,394 | -20% | 1 | 1 | 0% | 746 | 783 | +5% | 0 | 0 | — |
case-06 | pass→pass | 2,366 | 3,074 | +30% | 1 | 1 | 0% | 443 | 685 | +55% | 0 | 0 | — |
case-07 | pass→pass | 5,555 | 4,001 | -28% | 1 | 1 | 0% | 1,022 | 880 | -14% | 0 | 0 | — |
case-08 | pass→pass | 8,245 | 1,973 | -76% | 1 | 1 | 0% | 1,455 | 553 | -62% | 0 | 0 | — |
case-09 | fail→pass | 6,729 | 1,602 | -76% | 1 | 1 | 0% | 1,238 | 452 | -63% | 0 | 0 | — |
case-10 | fail→pass | 7,299 | 1,738 | -76% | 1 | 1 | 0% | 1,222 | 476 | -61% | 0 | 0 | — |
case-11 | pass→pass | 7,568 | 2,794 | -63% | 1 | 1 | 0% | 1,403 | 665 | -53% | 0 | 0 | — |
case-12 | fail→pass | 9,951 | 1,176 | -88% | 1 | 1 | 0% | 1,676 | 395 | -76% | 0 | 0 | — |
case-13 | pass→pass | 14,655 | 9,443 | -36% | 1 | 1 | 0% | 2,630 | 1,711 | -35% | 0 | 0 | — |
case-14 | fail→pass | 8,737 | 2,513 | -71% | 1 | 1 | 0% | 1,457 | 654 | -55% | 0 | 0 | — |
case-15 | pass→pass | 7,517 | 5,462 | -27% | 1 | 1 | 0% | 1,514 | 1,307 | -14% | 0 | 0 | — |
case-16 | fail→pass | 6,136 | 1,498 | -76% | 1 | 1 | 0% | 1,148 | 435 | -62% | 0 | 0 | — |
case-17 | fail→fail | 11,638 | 6,488 | -44% | 1 | 1 | 0% | 1,968 | 1,300 | -34% | 0 | 0 | — |
case-19 | pass→pass | 5,159 | 1,725 | -67% | 1 | 1 | 0% | 927 | 509 | -45% | 0 | 0 | — |
case-20 | pass→pass | 7,765 | 5,935 | -24% | 1 | 1 | 0% | 1,499 | 1,259 | -16% | 0 | 0 | — |
case-21 | pass→pass | 6,237 | 6,725 | +8% | 1 | 1 | 0% | 1,173 | 1,372 | +17% | 0 | 0 | — |
case-22 | pass→pass | 15,651 | 17,574 | +12% | 1 | 1 | 0% | 2,949 | 3,592 | +22% | 0 | 0 | — |
case-23 | pass→pass | 11,709 | 8,226 | -30% | 1 | 1 | 0% | 2,150 | 1,707 | -21% | 0 | 0 | — |
case-24 | pass→pass | 11,935 | 12,898 | +8% | 1 | 1 | 0% | 2,394 | 2,961 | +24% | 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. 24 cases were attempted. The headline lift of +42 percentage points is the difference between those two pass rates over the 24 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.