Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when targeting 《金融评论》(Chinese Review of Financial Studies — 中国社会科学院金融研究所主办、2009 年创刊、不收版面费的金融学术双月刊) or deciding whether a Chinese finance manuscript fits this venue. Encodes the journal's fit, framing, abstract/keyword/JEL house style, citation-quota rule, official ajcass.com submission re-check, and desk-reject heuristics.
.claude/skills/brycewang-stanford-chinese-review-of-financial-studies/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 50% | 0% |
《金融评论》由中国社会科学院金融研究所主办、中国社会科学院主管，2009 年创刊，是全国性金融学术刊物（CSSCI/北大核心）。覆盖金融理论与政策、资本市场、银行体系、公司金融、金融风险与金融开放。作为社科院系统刊物，重规范的理论框架与扎实识别，要求作者与既有文献充分对话。
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。
chinese-journal-of-management-science（《中国管理科学》） / chinese-public-administration（《中国行政管理》） / comparative-economic-and-social-systems（《经济社会体制比较》） / contemporary-economy-of-japan（《现代日本经济》）。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。经验识别强、央行系统取向转 journal-of-financial-research（《金融研究》）；证券市场微观结构与监管实务转 securities-market-herald（《证券市场导报》）；国际金融/汇率/跨境资本转 studies-of-international-finance（《国际金融研究》）。
text【匹配度】高 / 中 / 低（一句话理由） 【目标期刊】《金融评论》 【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 | 32,159 | 24,632 | -23% | 1 | 1 | 0% | 3,926 | 4,751 | +21% | 0 | 0 | — |
case-20 | pass→pass | 8,486 | 11,336 | +34% | 1 | 1 | 0% | 1,095 | 2,238 | +104% | 0 | 0 | — |
case-02 | fail→pass | 27,005 | 20,972 | -22% | 1 | 1 | 0% | 3,516 | 3,933 | +12% | 0 | 0 | — |
case-03 | fail→pass | 47,725 | 18,349 | -62% | 1 | 1 | 0% | 3,879 | 3,915 | +1% | 0 | 0 | — |
case-04 | pass→pass | 23,387 | 23,875 | +2% | 1 | 1 | 0% | 2,762 | 4,812 | +74% | 0 | 0 | — |
case-05 | pass→pass | 20,964 | 32,667 | +56% | 1 | 1 | 0% | 2,014 | 5,676 | +182% | 0 | 0 | — |
case-06 | pass→pass | 16,879 | 14,649 | -13% | 1 | 1 | 0% | 1,986 | 3,004 | +51% | 0 | 0 | — |
case-07 | pass→pass | 24,322 | 21,755 | -11% | 1 | 1 | 0% | 2,491 | 3,910 | +57% | 0 | 0 | — |
case-08 | fail→pass | 15,552 | 15,684 | +1% | 1 | 1 | 0% | 2,307 | 3,842 | +67% | 0 | 0 | — |
case-09 | pass→pass | 28,137 | 20,898 | -26% | 1 | 1 | 0% | 2,457 | 3,763 | +53% | 0 | 0 | — |
case-10 | fail→pass | 17,543 | 16,651 | -5% | 1 | 1 | 0% | 1,971 | 2,954 | +50% | 0 | 0 | — |
case-11 | fail→pass | 26,020 | 21,729 | -16% | 1 | 1 | 0% | 3,298 | 3,830 | +16% | 0 | 0 | — |
case-12 | fail→pass | 26,294 | 15,687 | -40% | 1 | 1 | 0% | 3,261 | 3,214 | -1% | 0 | 0 | — |
case-13 | pass→pass | 22,077 | 19,449 | -12% | 1 | 1 | 0% | 2,148 | 3,558 | +66% | 0 | 0 | — |
case-14 | pass→pass | 25,030 | 18,808 | -25% | 1 | 1 | 0% | 2,468 | 3,477 | +41% | 0 | 0 | — |
case-15 | fail→fail | 17,367 | 17,467 | +1% | 1 | 1 | 0% | 1,868 | 3,054 | +63% | 0 | 0 | — |
case-16 | fail→pass | 21,454 | 9,184 | -57% | 1 | 1 | 0% | 2,095 | 2,799 | +34% | 0 | 0 | — |
case-17 | pass→pass | 36,862 | 19,907 | -46% | 1 | 1 | 0% | 3,830 | 3,714 | -3% | 0 | 0 | — |
case-18 | pass→pass | 23,550 | 20,094 | -15% | 1 | 1 | 0% | 3,119 | 3,592 | +15% | 0 | 0 | — |
case-19 | pass→pass | 24,064 | 20,052 | -17% | 1 | 1 | 0% | 2,604 | 3,404 | +31% | 0 | 0 | — |
case-21 | pass→pass | 20,573 | 26,841 | +30% | 1 | 1 | 0% | 2,828 | 4,613 | +63% | 0 | 0 | — |
case-22 | pass→pass | 20,269 | 21,671 | +7% | 1 | 1 | 0% | 1,878 | 3,763 | +100% | 0 | 0 | — |
case-23 | fail→pass | 17,642 | 11,968 | -32% | 1 | 1 | 0% | 1,632 | 2,485 | +52% | 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 +39 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.