Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when targeting 《体育科学》(China Sport Science) — the China Sport Science Society flagship and top CSSCI sport-science journal (体育学). Apply to sport & exercise science spanning 运动人体科学 (生理/生化/生物力学/运动医学/体质健康), 体育人文社会学 (体育社会学/管理/经济/政策/史哲), 运动训练与竞技体育, 学校体育, 民族传统体育. Supports both natural-science experiments/interventions and social-science survey/qualitative/policy work; demands a real sport-science problem and discipline-appropriate rigor, not generic econometrics or pedagogy borrowed from other fiel
.claude/skills/brycewang-stanford-china-sport-science/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 17% | 0% |
中国体育科学学会主办，国内体育学第一刊（CSSCI、中文核心、AMI 权威、国家社科基金资助期刊）。它是一本综合性体育科学期刊，横跨运动人体科学（运动生理生化、生物力学、运动医学、体质与健康）与体育人文社会学（体育社会学、体育管理、体育经济、体育政策、体育史哲、学校体育、运动训练与竞技体育、民族传统体育）。核心要求是真问题 + 学科规范的证据：自然科学路径要实验/测试严谨、机制清楚；人文社科路径要问题意识、理论对话与方法规范。把别的学科（纯经济学政策评估、纯教育学经验总结）的稿子原样套进来，会显得"与体育学本体无关"。
cn-journal-workflow）；纯医学临床研究应回到临床医学渠道；纯教育学经验总结缺体育学机制的，不对口。正式建议前仍需进入官方核验清单。../../resources/source-basis.md 和 ../../resources/official-source-map.md，从其中的官方来源锚点或同一主办/出版体系入口开始核验，并在回答中说明核验日期。cn-journal-workflow 路由）【匹配度】高 / 中 / 低
【路径】运动人体科学(实验/测试) / 体育人文社会学(定量/定性/政策) / 跨学科混合
【问题意识】<体育学本体问题一句话>
【学科对话】<运动训练学/运动生理/体育社会科学理论> | 缺失
【机制解释】运动学/体育治理含义清晰 / 仅系数或指标（需补）
【贡献】推进/修正 <体育科学认识>| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 30,273 | 20,092 | -34% | 1 | 1 | 0% | 3,082 | 3,597 | +17% | 0 | 0 | — |
case-14 | pass→pass | 18,438 | 12,796 | -31% | 1 | 1 | 0% | 2,596 | 3,693 | +42% | 0 | 0 | — |
case-09 | fail→pass | 27,292 | 25,206 | -8% | 1 | 1 | 0% | 3,108 | 4,357 | +40% | 0 | 0 | — |
case-02 | fail→pass | 36,853 | 25,499 | -31% | 1 | 1 | 0% | 3,658 | 4,465 | +22% | 0 | 0 | — |
case-03 | fail→pass | 34,551 | 25,982 | -25% | 1 | 1 | 0% | 3,537 | 5,129 | +45% | 0 | 0 | — |
case-04 | fail→fail | 34,271 | 25,647 | -25% | 1 | 1 | 0% | 3,897 | 4,208 | +8% | 0 | 0 | — |
case-05 | pass→pass | 25,364 | 32,392 | +28% | 1 | 1 | 0% | 3,943 | 5,680 | +44% | 0 | 0 | — |
case-06 | fail→pass | 33,418 | 26,722 | -20% | 1 | 1 | 0% | 4,463 | 5,234 | +17% | 0 | 0 | — |
case-07 | pass→pass | 28,832 | 25,974 | -10% | 1 | 1 | 0% | 3,044 | 4,281 | +41% | 0 | 0 | — |
case-08 | fail→pass | 25,907 | 19,093 | -26% | 1 | 1 | 0% | 2,985 | 3,687 | +24% | 0 | 0 | — |
case-10 | fail→fail | 27,087 | 28,976 | +7% | 1 | 1 | 0% | 3,069 | 4,661 | +52% | 0 | 0 | — |
case-11 | pass→pass | 25,562 | 19,288 | -25% | 1 | 1 | 0% | 2,527 | 4,354 | +72% | 0 | 0 | — |
case-12 | pass→pass | 25,963 | 16,601 | -36% | 1 | 1 | 0% | 2,628 | 3,834 | +46% | 0 | 0 | — |
case-13 | fail→pass | 27,911 | 20,218 | -28% | 1 | 1 | 0% | 3,045 | 4,238 | +39% | 0 | 0 | — |
case-15 | pass→pass | 28,312 | 23,498 | -17% | 1 | 1 | 0% | 3,058 | 5,008 | +64% | 0 | 0 | — |
case-16 | fail→pass | 25,629 | 21,882 | -15% | 1 | 1 | 0% | 3,222 | 3,828 | +19% | 0 | 0 | — |
case-17 | fail→fail | 25,641 | 25,786 | +1% | 1 | 1 | 0% | 2,623 | 4,476 | +71% | 0 | 0 | — |
case-18 | pass→pass | 22,398 | 28,801 | +29% | 1 | 1 | 0% | 3,211 | 4,756 | +48% | 0 | 0 | — |
case-19 | pass→pass | 23,984 | 13,672 | -43% | 1 | 1 | 0% | 2,607 | 3,651 | +40% | 0 | 0 | — |
case-20 | pass→pass | 23,295 | 26,263 | +13% | 1 | 1 | 0% | 2,468 | 4,538 | +84% | 0 | 0 | — |
case-21 | fail→pass | 21,036 | 22,757 | +8% | 1 | 1 | 0% | 2,963 | 4,126 | +39% | 0 | 0 | — |
case-22 | pass→pass | 26,552 | 29,189 | +10% | 1 | 1 | 0% | 3,132 | 4,694 | +50% | 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 +41 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.