Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Create product backlog items in Why-What-Acceptance format — independent, valuable, testable items with strategic context. Use when writing structured backlog items, breaking features into work items, or using the WWA format.
.claude/skills/phuryn-wwas/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 19% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 26% | 0% |
Create product backlog items in Why-What-Acceptance format. Produces independent, valuable, testable items with strategic context.
Use when: Writing backlog items, creating product increments, breaking features into work items, or communicating strategic intent to teams.
Arguments:
$PRODUCT: The product or system name$FEATURE: The new feature or capability$DESIGN: Link to design files (Figma, Miro, etc.)$ASSUMPTIONS: Key assumptions and strategic contextTitle: What will be delivered]
Why: 1-2 sentences connecting to strategic context and team objectives]
What: Short description and design link. 1-2 paragraphs maximum. A reminder of discussion, not detailed specification.]
Acceptance Criteria:
Title: Implement Real-Time Spending Tracker
Why: Users need immediate feedback on spending to make conscious budget decisions. This directly supports our goal to improve financial awareness and reduce overspending.
What: Add a real-time spending tracker that updates as users log expenses. The tracker displays their current week's spending against their set budget. Designs available in Figma link]. This is a reminder of our discussions - detailed specifications will emerge during development conversations with the team.
Acceptance Criteria:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 14,159 | 13,451 | -5% | 1 | 1 | 0% | 2,570 | 2,920 | +14% | 0 | 0 | — |
case-02 | fail→pass | 17,171 | 12,147 | -29% | 1 | 1 | 0% | 3,116 | 2,819 | -10% | 0 | 0 | — |
case-03 | pass→fail | 16,902 | 19,159 | +13% | 1 | 1 | 0% | 2,739 | 3,462 | +26% | 0 | 0 | — |
case-04 | pass→fail | 14,947 | 17,206 | +15% | 1 | 1 | 0% | 2,470 | 3,463 | +40% | 0 | 0 | — |
case-05 | pass→pass | 17,620 | 13,759 | -22% | 1 | 1 | 0% | 2,897 | 3,029 | +5% | 0 | 0 | — |
case-06 | fail→pass | 8,671 | 7,584 | -13% | 1 | 1 | 0% | 1,655 | 1,970 | +19% | 0 | 0 | — |
case-07 | pass→pass | 11,769 | 4,817 | -59% | 1 | 1 | 0% | 1,535 | 1,460 | -5% | 0 | 0 | — |
case-08 | fail→pass | 9,558 | 5,685 | -41% | 1 | 1 | 0% | 1,484 | 1,646 | +11% | 0 | 0 | — |
case-09 | pass→fail | 9,924 | 7,087 | -29% | 1 | 1 | 0% | 1,676 | 1,504 | -10% | 0 | 0 | — |
case-10 | fail→fail | 11,008 | 6,465 | -41% | 1 | 1 | 0% | 2,003 | 1,755 | -12% | 0 | 0 | — |
case-11 | fail→fail | 7,551 | 5,375 | -29% | 1 | 1 | 0% | 1,397 | 1,583 | +13% | 0 | 0 | — |
case-12 | pass→pass | 8,643 | 5,907 | -32% | 1 | 1 | 0% | 1,150 | 1,502 | +31% | 0 | 0 | — |
case-13 | pass→pass | 10,738 | 7,479 | -30% | 1 | 1 | 0% | 1,515 | 1,669 | +10% | 0 | 0 | — |
case-14 | fail→fail | 13,555 | 6,027 | -56% | 1 | 1 | 0% | 1,964 | 1,715 | -13% | 0 | 0 | — |
case-15 | fail→fail | 11,077 | 9,308 | -16% | 1 | 1 | 0% | 1,865 | 1,724 | -8% | 0 | 0 | — |
case-16 | fail→fail | 8,964 | 6,742 | -25% | 1 | 1 | 0% | 1,649 | 1,789 | +8% | 0 | 0 | — |
case-17 | fail→fail | 14,523 | 9,582 | -34% | 1 | 1 | 0% | 2,120 | 2,018 | -5% | 0 | 0 | — |
case-18 | pass→pass | 7,420 | 6,069 | -18% | 1 | 1 | 0% | 1,351 | 1,459 | +8% | 0 | 0 | — |
case-19 | fail→pass | 10,886 | 6,877 | -37% | 1 | 1 | 0% | 1,970 | 1,823 | -7% | 0 | 0 | — |
case-20 | pass→pass | 8,523 | 5,117 | -40% | 1 | 1 | 0% | 1,616 | 1,543 | -5% | 0 | 0 | — |
case-21 | pass→pass | 7,678 | 6,153 | -20% | 1 | 1 | 0% | 1,413 | 1,458 | +3% | 0 | 0 | — |
case-22 | fail→fail | 5,515 | 11,057 | +100% | 1 | 1 | 0% | 1,062 | 1,875 | +77% | 0 | 0 | — |
case-23 | fail→fail | 9,386 | 5,539 | -41% | 1 | 1 | 0% | 1,198 | 1,444 | +21% | 0 | 0 | — |
case-24 | fail→fail | 9,313 | 6,820 | -27% | 1 | 1 | 0% | 1,582 | 2,015 | +27% | 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 +4 percentage points is the difference between those two pass rates over the 24 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.