Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Migrate test files from `as` type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace `as` in tests, or needs partial test data.
.claude/skills/mkurman-migrate-to-shoehorn/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -67% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -51% | 0% |
--------- | -------------------------------------------------- | | fromPartial() | Pass partial data that still type-checks | | fromAny() | Pass intentionally wrong data (keeps autocomplete) | | fromExact() | Force full object (swap with fromPartial later) |
as assertions causing problems?npm i @total-typescript/shoehornas assertions: grep -r " as [A-Z]" --include="*.test.ts" --include="*.spec.ts"as Type with fromPartial()as unknown as Type with fromAny()@total-typescript/shoehorn| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 17,070 | 12,020 | -30% | 1 | 1 | 0% | 3,031 | 2,373 | -22% | 0 | 0 | — |
case-02 | fail→fail | 16,939 | 7,003 | -59% | 1 | 1 | 0% | 3,046 | 1,488 | -51% | 0 | 0 | — |
case-03 | fail→pass | 14,561 | 6,255 | -57% | 1 | 1 | 0% | 2,694 | 1,298 | -52% | 0 | 0 | — |
case-04 | pass→pass | 10,232 | 5,856 | -43% | 1 | 1 | 0% | 2,396 | 1,301 | -46% | 0 | 0 | — |
case-05 | pass→pass | 10,736 | 7,753 | -28% | 1 | 1 | 0% | 2,209 | 1,667 | -25% | 0 | 0 | — |
case-06 | pass→pass | 8,634 | 6,352 | -26% | 1 | 1 | 0% | 1,496 | 1,311 | -12% | 0 | 0 | — |
case-07 | fail→pass | 14,131 | 3,489 | -75% | 1 | 1 | 0% | 2,586 | 843 | -67% | 0 | 0 | — |
case-08 | fail→pass | 6,708 | 2,842 | -58% | 1 | 1 | 0% | 1,112 | 744 | -33% | 0 | 0 | — |
case-09 | fail→pass | 9,531 | 3,643 | -62% | 1 | 1 | 0% | 1,723 | 872 | -49% | 0 | 0 | — |
case-10 | pass→pass | 12,033 | 4,736 | -61% | 1 | 1 | 0% | 2,139 | 1,065 | -50% | 0 | 0 | — |
case-11 | fail→pass | 9,711 | 2,917 | -70% | 1 | 1 | 0% | 1,524 | 740 | -51% | 0 | 0 | — |
case-12 | fail→pass | 9,068 | 1,792 | -80% | 1 | 1 | 0% | 1,410 | 456 | -68% | 0 | 0 | — |
case-13 | pass→pass | 5,685 | 2,739 | -52% | 1 | 1 | 0% | 911 | 689 | -24% | 0 | 0 | — |
case-14 | fail→pass | 11,825 | 4,954 | -58% | 1 | 1 | 0% | 2,094 | 1,088 | -48% | 0 | 0 | — |
case-15 | fail→pass | 12,667 | 2,787 | -78% | 1 | 1 | 0% | 2,061 | 718 | -65% | 0 | 0 | — |
case-16 | pass→pass | 14,299 | 3,457 | -76% | 1 | 1 | 0% | 2,379 | 825 | -65% | 0 | 0 | — |
case-17 | pass→pass | 7,951 | 3,309 | -58% | 1 | 1 | 0% | 1,328 | 826 | -38% | 0 | 0 | — |
case-18 | fail→pass | 11,688 | 3,980 | -66% | 1 | 1 | 0% | 1,993 | 978 | -51% | 0 | 0 | — |
case-19 | fail→pass | 15,773 | 1,695 | -89% | 1 | 1 | 0% | 2,547 | 488 | -81% | 0 | 0 | — |
case-20 | fail→fail | 7,990 | 4,631 | -42% | 1 | 1 | 0% | 1,304 | 1,000 | -23% | 0 | 0 | — |
case-21 | fail→pass | 15,736 | 3,592 | -77% | 1 | 1 | 0% | 2,681 | 822 | -69% | 0 | 0 | — |
case-22 | fail→fail | 12,998 | 8,404 | -35% | 1 | 1 | 0% | 2,296 | 1,756 | -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. 22 cases were attempted. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 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.