Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when targeting 《中国会计评论》(China Accounting Review — 北京大学等高校发起、依托北京大学光华管理学院、实行匿名审稿的会计学术季刊) or deciding whether a Chinese accounting/finance manuscript fits this venue. Encodes the journal's fit, framing, anonymous-review house style, official-submission re-check, and desk-reject heuristics.
.claude/skills/brycewang-stanford-china-accounting-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 21% | 0% |
《中国会计评论》（China Accounting Review）是由北京大学、清华大学等高校发起、依托北京大学光华管理学院运作的会计学术季刊，2003 年创刊。定位偏学术型、规范化的会计研究，覆盖财务会计、资本市场会计、审计与披露、会计准则与公司治理，方法上以严谨的档案实证、理论与实验研究为主。相对实务/政策导向刊物，本刊更看重会计构念的清晰构造、识别策略与对会计文献的增量贡献，强调与国际规范接轨。
这个 skill 是定位 / 选刊 / 改写框架工具，不替代该刊最新官方投稿须知。正式投稿前必须重新核对官网、采编系统或编辑部发布的最新模板。
bulletin-of-chinese-academy-of-sciences（《中国科学院院刊》） / business-management-journal（《经济管理》） / china-economic-quarterly（《经济学（季刊）》） / china-economic-studies（《中国经济问题》）。若这些刊物的读者对象更贴近，不要因为名称、地区或变量相似而强行投本刊。../../resources/source-basis.md 和 ../../resources/official-source-map.md，从官方来源锚点开始核验，并在回答中说明核验日期。中文会计旗舰转 accounting-research（《会计研究》）；审计专业稿转 auditing-research（《审计研究》）；会计与经济交叉、门槛略低转 accounting-and-economics-research（《会计与经济研究》）。
text【匹配度】高 / 中 / 低（一句话理由） 【目标期刊】《中国会计评论》 【选题标签】<最贴近的 2-3 个主题> 【方法证据】<当前方法是否够本刊标准> 【最大风险】<最可能导致退稿的一点> 【需核验官方要求】<投稿系统/匿名/摘要/图表/数据等> 【改投建议】<若不匹配，给出更合适期刊>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 33,421 | 22,615 | -32% | 1 | 1 | 0% | 3,588 | 3,772 | +5% | 0 | 0 | — |
case-17 | fail→fail | 31,909 | 21,251 | -33% | 1 | 1 | 0% | 3,608 | 3,375 | -6% | 0 | 0 | — |
case-23 | fail→fail | 13,902 | 15,547 | +12% | 1 | 1 | 0% | 1,107 | 3,035 | +174% | 0 | 0 | — |
case-02 | fail→pass | 34,258 | 21,975 | -36% | 1 | 1 | 0% | 3,861 | 3,940 | +2% | 0 | 0 | — |
case-03 | fail→pass | 35,196 | 13,458 | -62% | 1 | 1 | 0% | 3,837 | 3,361 | -12% | 0 | 0 | — |
case-04 | fail→pass | 23,873 | 20,470 | -14% | 1 | 1 | 0% | 2,993 | 3,858 | +29% | 0 | 0 | — |
case-05 | fail→pass | 25,411 | 28,762 | +13% | 1 | 1 | 0% | 3,350 | 4,067 | +21% | 0 | 0 | — |
case-06 | fail→pass | 26,490 | 29,579 | +12% | 1 | 1 | 0% | 3,038 | 4,891 | +61% | 0 | 0 | — |
case-07 | fail→pass | 30,086 | 19,437 | -35% | 1 | 1 | 0% | 3,294 | 3,456 | +5% | 0 | 0 | — |
case-08 | fail→fail | 27,043 | 25,047 | -7% | 1 | 1 | 0% | 2,776 | 4,155 | +50% | 0 | 0 | — |
case-09 | fail→pass | 25,545 | 33,351 | +31% | 1 | 1 | 0% | 2,977 | 5,064 | +70% | 0 | 0 | — |
case-10 | fail→fail | 33,046 | 19,179 | -42% | 1 | 1 | 0% | 3,387 | 4,294 | +27% | 0 | 0 | — |
case-11 | fail→fail | 23,050 | 17,171 | -26% | 1 | 1 | 0% | 2,864 | 3,614 | +26% | 0 | 0 | — |
case-12 | fail→fail | 32,573 | 30,814 | -5% | 1 | 1 | 0% | 3,689 | 4,939 | +34% | 0 | 0 | — |
case-13 | fail→fail | 28,530 | 21,478 | -25% | 1 | 1 | 0% | 2,889 | 4,237 | +47% | 0 | 0 | — |
case-14 | pass→pass | 26,098 | 24,040 | -8% | 1 | 1 | 0% | 2,683 | 3,844 | +43% | 0 | 0 | — |
case-15 | fail→pass | 37,624 | 23,995 | -36% | 1 | 1 | 0% | 3,883 | 3,724 | -4% | 0 | 0 | — |
case-16 | fail→fail | 56,504 | 28,309 | -50% | 1 | 1 | 0% | 2,936 | 4,743 | +62% | 0 | 0 | — |
case-18 | fail→pass | 30,756 | 25,730 | -16% | 1 | 1 | 0% | 3,967 | 4,138 | +4% | 0 | 0 | — |
case-19 | fail→pass | 32,383 | 28,087 | -13% | 1 | 1 | 0% | 3,586 | 4,620 | +29% | 0 | 0 | — |
case-20 | pass→pass | 27,617 | 31,272 | +13% | 1 | 1 | 0% | 2,933 | 4,782 | +63% | 0 | 0 | — |
case-21 | pass→pass | 23,243 | 30,846 | +33% | 1 | 1 | 0% | 2,642 | 4,650 | +76% | 0 | 0 | — |
case-22 | pass→pass | 27,558 | 36,586 | +33% | 1 | 1 | 0% | 4,819 | 7,005 | +45% | 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. 23 cases were attempted. The headline lift of +48 percentage points is the difference between those two pass rates over the 23 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.