Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze and prioritize a list of feature requests by theme, strategic alignment, impact, effort, and risk. Use when reviewing customer feature requests, triaging a backlog, or making prioritization decisions.
.claude/skills/phuryn-analyze-feature-requests/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -17% | 0% |
Categorize, evaluate, and prioritize customer feature requests against product goals.
You are analyzing feature requests for $ARGUMENTS.
If the user provides files (spreadsheets, CSVs, or documents with feature requests), read and analyze them directly. If data is in a structured format, consider creating a summary table.
Never allow customers to design solutions. Prioritize opportunities (problems), not features. Use Opportunity Score (Dan Olsen) to evaluate customer-reported problems: Opportunity Score = Importance × (1 − Satisfaction), normalized to 0–1. See the prioritization-frameworks skill for full details and templates.
The user will describe their product goal and provide feature requests. Work through these steps:
Think step by step. Save as markdown or create a structured output document.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→fail | 11,650 | 18,866 | +62% | 1 | 1 | 0% | 2,208 | 3,444 | +56% | 0 | 0 | — |
case-01 | fail→pass | 20,716 | 20,808 | +0% | 1 | 1 | 0% | 3,437 | 4,126 | +20% | 0 | 0 | — |
case-02 | fail→fail | 21,383 | 25,973 | +21% | 1 | 1 | 0% | 3,577 | 3,838 | +7% | 0 | 0 | — |
case-03 | fail→fail | 17,480 | 26,294 | +50% | 1 | 1 | 0% | 3,195 | 5,012 | +57% | 0 | 0 | — |
case-04 | pass→pass | 11,425 | 8,533 | -25% | 1 | 1 | 0% | 2,237 | 2,183 | -2% | 0 | 0 | — |
case-05 | pass→pass | 22,613 | 18,540 | -18% | 1 | 1 | 0% | 3,797 | 3,821 | +1% | 0 | 0 | — |
case-06 | pass→pass | 10,504 | 11,464 | +9% | 1 | 1 | 0% | 1,683 | 2,279 | +35% | 0 | 0 | — |
case-07 | fail→pass | 6,862 | 7,014 | +2% | 1 | 1 | 0% | 1,352 | 1,365 | +1% | 0 | 0 | — |
case-08 | pass→pass | 13,285 | 29,365 | +121% | 1 | 1 | 0% | 2,427 | 3,733 | +54% | 0 | 0 | — |
case-09 | fail→pass | 10,433 | 4,105 | -61% | 1 | 1 | 0% | 1,507 | 1,121 | -26% | 0 | 0 | — |
case-10 | pass→pass | 15,336 | 14,798 | -4% | 1 | 1 | 0% | 2,475 | 2,898 | +17% | 0 | 0 | — |
case-11 | fail→pass | 6,578 | 2,634 | -60% | 1 | 1 | 0% | 1,216 | 911 | -25% | 0 | 0 | — |
case-12 | pass→pass | 10,436 | 12,704 | +22% | 1 | 1 | 0% | 1,841 | 2,486 | +35% | 0 | 0 | — |
case-13 | fail→fail | 8,830 | 11,367 | +29% | 1 | 1 | 0% | 1,408 | 1,717 | +22% | 0 | 0 | — |
case-14 | pass→pass | 7,223 | 2,371 | -67% | 1 | 1 | 0% | 1,524 | 882 | -42% | 0 | 0 | — |
case-15 | fail→pass | 6,152 | 3,807 | -38% | 1 | 1 | 0% | 1,179 | 975 | -17% | 0 | 0 | — |
case-16 | fail→pass | 7,490 | 3,753 | -50% | 1 | 1 | 0% | 1,168 | 1,104 | -5% | 0 | 0 | — |
case-17 | fail→pass | 14,684 | 21,798 | +48% | 1 | 1 | 0% | 2,555 | 4,213 | +65% | 0 | 0 | — |
case-18 | pass→pass | 11,173 | 10,398 | -7% | 1 | 1 | 0% | 1,979 | 2,201 | +11% | 0 | 0 | — |
case-19 | fail→pass | 5,949 | 3,709 | -38% | 1 | 1 | 0% | 1,210 | 1,001 | -17% | 0 | 0 | — |
case-20 | pass→fail | 38,017 | 31,432 | -17% | 1 | 1 | 0% | 5,543 | 6,606 | +19% | 0 | 0 | — |
case-22 | pass→pass | 12,079 | 14,546 | +20% | 1 | 1 | 0% | 2,290 | 2,422 | +6% | 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 +27 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 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.