Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Create a structured user interview script with warm-up, core exploration, and wrap-up sections. Use when preparing for user research interviews to ensure consistent, insightful conversations.
.claude/skills/owl-listener-interview-script/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-09 | ✓→✗ | ▼ Worse | 41% | 0% |
| case-11 | ✓→✗ | ▼ Worse | 73% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 6% | 0% |
Create a structured user interview script for qualitative research.
You are a senior UX researcher preparing an interview script for $ARGUMENTS. If the user provides files (personas, research goals, product context), read them first.
Ensure all questions are non-leading. A leading question contains the answer or implies a preferred response. Replace any question that assumes sentiment, behaviour, or outcome.
| Leading (avoid) | Non-leading (use) | |---|---| | "Was that frustrating?" | "How did you feel about that?" | | "Did you find it easy?" | "How easy or difficult was that for you?" | | "Did you like the feature?" | "What did you notice about that feature, if anything?" | | "Would you use this?" | "How would you use this in your work, if at all?" |
Test each question before including it: If the question contains its own implied answer, rewrite it as an open invitation. If the question can be answered with yes or no, extend it ("...and why?").
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 22,203 | 18,457 | -17% | 1 | 1 | 0% | 3,657 | 4,151 | +14% | 0 | 0 | — |
case-02 | fail→fail | 16,419 | 18,482 | +13% | 1 | 1 | 0% | 2,837 | 3,560 | +25% | 0 | 0 | — |
case-03 | pass→pass | 17,109 | 15,655 | -8% | 1 | 1 | 0% | 3,078 | 3,276 | +6% | 0 | 0 | — |
case-04 | fail→fail | 15,011 | 13,786 | -8% | 1 | 1 | 0% | 2,584 | 3,143 | +22% | 0 | 0 | — |
case-05 | pass→pass | 12,777 | 15,963 | +25% | 1 | 1 | 0% | 1,945 | 3,370 | +73% | 0 | 0 | — |
case-06 | pass→pass | 11,225 | 17,422 | +55% | 1 | 1 | 0% | 1,795 | 3,643 | +103% | 0 | 0 | — |
case-07 | pass→pass | 10,239 | 14,495 | +42% | 1 | 1 | 0% | 1,707 | 2,828 | +66% | 0 | 0 | — |
case-08 | pass→pass | 11,554 | 18,153 | +57% | 1 | 1 | 0% | 1,914 | 3,498 | +83% | 0 | 0 | — |
case-09 | pass→fail | 12,981 | 16,533 | +27% | 1 | 1 | 0% | 2,421 | 3,413 | +41% | 0 | 0 | — |
case-20 | pass→pass | 9,126 | 14,597 | +60% | 1 | 1 | 0% | 1,783 | 3,180 | +78% | 0 | 0 | — |
case-10 | pass→pass | 10,273 | 7,541 | -27% | 1 | 1 | 0% | 1,739 | 1,738 | -0% | 0 | 0 | — |
case-11 | pass→fail | 16,661 | 22,240 | +33% | 1 | 1 | 0% | 2,489 | 4,311 | +73% | 0 | 0 | — |
case-12 | fail→pass | 17,948 | 13,264 | -26% | 1 | 1 | 0% | 2,930 | 2,658 | -9% | 0 | 0 | — |
case-13 | fail→fail | 13,210 | 17,513 | +33% | 1 | 1 | 0% | 2,216 | 3,505 | +58% | 0 | 0 | — |
case-21 | pass→pass | 20,320 | 18,982 | -7% | 1 | 1 | 0% | 3,442 | 3,746 | +9% | 0 | 0 | — |
case-14 | fail→pass | 10,469 | 7,282 | -30% | 1 | 1 | 0% | 1,844 | 1,871 | +1% | 0 | 0 | — |
case-15 | pass→pass | 10,492 | 7,693 | -27% | 1 | 1 | 0% | 1,584 | 1,797 | +13% | 0 | 0 | — |
case-16 | pass→pass | 11,978 | 14,843 | +24% | 1 | 1 | 0% | 2,050 | 3,255 | +59% | 0 | 0 | — |
case-17 | pass→pass | 12,632 | 14,227 | +13% | 1 | 1 | 0% | 1,962 | 2,995 | +53% | 0 | 0 | — |
case-18 | pass→pass | 10,723 | 16,429 | +53% | 1 | 1 | 0% | 1,751 | 3,161 | +81% | 0 | 0 | — |
case-19 | pass→pass | 11,595 | 8,497 | -27% | 1 | 1 | 0% | 1,872 | 1,956 | +4% | 0 | 0 | — |
case-22 | pass→pass | 18,031 | 16,767 | -7% | 1 | 1 | 0% | 2,863 | 3,450 | +21% | 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 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are 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.