Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when preparing the narrative, sources, and design context for a new editable PPTX deck.
.claude/skills/wshobson-pptx-deck-context/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 51% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -6% | 0% |
Prepare the deck before coordinates or PPTX objects are authored. This skill owns the business narrative, source lineage, and design lock.
mckinsey, scqa, pyramid, mece, action-title, assertion-evidence, exec-summary-first, or custom; do not choose silently.source_ref for every metric, quotation, chart value, and factual claim.references/design-profiles.md.summary before authoring slides.See references/design-profiles.md for reusable profile guidance.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 7,849 | 4,212 | -46% | 1 | 1 | 0% | 1,493 | 1,232 | -17% | 0 | 0 | — |
case-01 | fail→pass | 74,726 | 10,942 | -85% | 1 | 1 | 0% | 2,185 | 2,396 | +10% | 0 | 0 | — |
case-02 | fail→pass | 14,637 | 11,328 | -23% | 1 | 1 | 0% | 2,857 | 2,513 | -12% | 0 | 0 | — |
case-03 | fail→pass | 15,170 | 6,083 | -60% | 1 | 1 | 0% | 2,896 | 1,473 | -49% | 0 | 0 | — |
case-05 | pass→fail | 13,734 | 11,415 | -17% | 1 | 1 | 0% | 2,330 | 2,548 | +9% | 0 | 0 | — |
case-06 | pass→pass | 13,463 | 6,086 | -55% | 1 | 1 | 0% | 2,534 | 1,632 | -36% | 0 | 0 | — |
case-07 | pass→pass | 12,644 | 8,735 | -31% | 1 | 1 | 0% | 2,196 | 1,963 | -11% | 0 | 0 | — |
case-08 | pass→pass | 10,175 | 4,312 | -58% | 1 | 1 | 0% | 1,444 | 1,181 | -18% | 0 | 0 | — |
case-09 | fail→pass | 7,690 | 5,907 | -23% | 1 | 1 | 0% | 1,033 | 1,564 | +51% | 0 | 0 | — |
case-10 | fail→pass | 8,746 | 7,047 | -19% | 1 | 1 | 0% | 1,651 | 1,554 | -6% | 0 | 0 | — |
case-11 | pass→pass | 7,214 | 2,459 | -66% | 1 | 1 | 0% | 1,354 | 869 | -36% | 0 | 0 | — |
case-12 | fail→pass | 10,327 | 6,373 | -38% | 1 | 1 | 0% | 1,883 | 1,597 | -15% | 0 | 0 | — |
case-13 | pass→pass | 11,768 | 5,144 | -56% | 1 | 1 | 0% | 2,135 | 1,356 | -36% | 0 | 0 | — |
case-14 | pass→pass | 8,564 | 9,074 | +6% | 1 | 1 | 0% | 1,604 | 1,943 | +21% | 0 | 0 | — |
case-15 | fail→pass | 10,449 | 3,607 | -65% | 1 | 1 | 0% | 1,776 | 865 | -51% | 0 | 0 | — |
case-16 | fail→pass | 11,557 | 6,231 | -46% | 1 | 1 | 0% | 1,863 | 1,496 | -20% | 0 | 0 | — |
case-17 | fail→pass | 11,227 | 3,061 | -73% | 1 | 1 | 0% | 1,991 | 852 | -57% | 0 | 0 | — |
case-18 | pass→pass | 10,233 | 4,338 | -58% | 1 | 1 | 0% | 1,703 | 1,085 | -36% | 0 | 0 | — |
case-19 | pass→pass | 8,795 | 5,179 | -41% | 1 | 1 | 0% | 1,442 | 1,247 | -14% | 0 | 0 | — |
case-20 | pass→pass | 4,760 | 10,875 | +128% | 1 | 1 | 0% | 1,135 | 3,157 | +178% | 0 | 0 | — |
case-21 | pass→pass | 12,816 | 10,481 | -18% | 1 | 1 | 0% | 2,537 | 2,418 | -5% | 0 | 0 | — |
case-22 | pass→fail | 69,443 | 4,768 | -93% | 1 | 1 | 0% | 1,652 | 1,109 | -33% | 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 +32 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.