Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →同步中文毕业论文的非正文部分：摘要、附录、封面、成果、致谢、名单等，确保这些部分与正文口径一致，而不是最后临时补写。 **Trigger**: 摘要同步, frontmatter sync, 附录同步, 封面检查, 非正文维护, 毕业论文摘要. **Use when**: 正文结构已经比较稳定，需要并行维护非正文部分，避免中英文摘要、封面字段、附录和成果列表与正文脱节。 **Skip if**: 正文主线还不稳定，或者还没有任何可交付版本。 **Network**: none. **Guardrail**: 不把非正文拖到最后一刻；不让摘要、附录与正文口径脱节。
.claude/skills/willoscar-thesis-frontmatter-sync/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -60% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -79% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -17% | 0% |
abstract/abstract.texabstract/abstract-en.texpreface/...chapters/appendix.texacknowledgement/...achievements/...Always read:
references/overview.mdreferences/frontmatter_checklist.mdMachine-readable contract:
assets/frontmatter_sync_contract.json| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 40,551 | 9,442 | -77% | 1 | 1 | 0% | 2,123 | 1,722 | -19% | 0 | 0 | — |
case-21 | pass→pass | 15,268 | 15,797 | +3% | 1 | 1 | 0% | 2,910 | 1,912 | -34% | 0 | 0 | — |
case-05 | fail→pass | 8,047 | 2,544 | -68% | 1 | 1 | 0% | 1,353 | 613 | -55% | 0 | 0 | — |
case-01 | fail→fail | 19,618 | 14,907 | -24% | 1 | 1 | 0% | 3,255 | 2,492 | -23% | 0 | 0 | — |
case-02 | fail→fail | 13,960 | 4,500 | -68% | 1 | 1 | 0% | 2,364 | 862 | -64% | 0 | 0 | — |
case-03 | fail→fail | 13,720 | 9,896 | -28% | 1 | 1 | 0% | 2,128 | 1,823 | -14% | 0 | 0 | — |
case-04 | pass→pass | 10,345 | 2,587 | -75% | 1 | 1 | 0% | 1,623 | 527 | -68% | 0 | 0 | — |
case-06 | fail→pass | 7,239 | 1,679 | -77% | 1 | 1 | 0% | 1,063 | 440 | -59% | 0 | 0 | — |
case-07 | fail→pass | 6,039 | 1,895 | -69% | 1 | 1 | 0% | 1,038 | 416 | -60% | 0 | 0 | — |
case-08 | fail→pass | 10,720 | 1,441 | -87% | 1 | 1 | 0% | 1,628 | 345 | -79% | 0 | 0 | — |
case-09 | fail→pass | 5,305 | 2,456 | -54% | 1 | 1 | 0% | 686 | 571 | -17% | 0 | 0 | — |
case-10 | fail→fail | 6,803 | 1,694 | -75% | 1 | 1 | 0% | 978 | 398 | -59% | 0 | 0 | — |
case-11 | fail→pass | 2,320 | 4,589 | +98% | 1 | 1 | 0% | 291 | 851 | +192% | 0 | 0 | — |
case-12 | fail→pass | 3,776 | 3,861 | +2% | 1 | 1 | 0% | 523 | 791 | +51% | 0 | 0 | — |
case-13 | fail→pass | 5,363 | 1,890 | -65% | 1 | 1 | 0% | 706 | 361 | -49% | 0 | 0 | — |
case-14 | fail→pass | 9,582 | 2,123 | -78% | 1 | 1 | 0% | 1,524 | 527 | -65% | 0 | 0 | — |
case-15 | fail→pass | 9,857 | 1,830 | -81% | 1 | 1 | 0% | 1,461 | 414 | -72% | 0 | 0 | — |
case-16 | pass→pass | 6,489 | 2,319 | -64% | 1 | 1 | 0% | 1,005 | 496 | -51% | 0 | 0 | — |
case-17 | fail→pass | 11,218 | 8,842 | -21% | 1 | 1 | 0% | 1,758 | 1,551 | -12% | 0 | 0 | — |
case-18 | fail→pass | 11,350 | 18,899 | +67% | 1 | 1 | 0% | 1,641 | 1,319 | -20% | 0 | 0 | — |
case-19 | pass→pass | 40,261 | 9,465 | -76% | 1 | 1 | 0% | 1,765 | 697 | -61% | 0 | 0 | — |
case-22 | pass→pass | 9,235 | 9,455 | +2% | 1 | 1 | 0% | 1,469 | 1,578 | +7% | 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 +55 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.