Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when targeting 《经济学动态》(Economic Perspectives — 中国社会科学院主管、经济研究所主办的理论经济学月刊, 1960 年创刊, CSSCI) or deciding whether a Chinese econ manuscript fits this venue. Encodes the journal's fit, framing, abstract/keyword house style, official-submission re-check, and desk-reject heuristics.
.claude/skills/brycewang-stanford-economic-perspectives/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 52% | 0% |
《经济学动态》由中国社会科学院主管、中国社会科学院经济研究所主办，1960 年创刊，是理论经济学月刊，CSSCI 来源期刊。是经济学综述、学术动态与前沿评论的重要刊物，适合把新理论、新方法或中国经济重大问题梳理成规范学术问题——经济学前沿综述与争鸣、重大问题理论评论、政策热点的学术化评述与研究议程。强调理论深度、独到见解与新意。
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。
east-china-economic-management（《华东经济管理》） / economic-aspects（《经济纵横》） / economic-problems（《经济问题》） / economic-research（《经济研究》）。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。原创实证论文更适合经济研究/CEQ/中国工业经济；纯政策评论需增强文献和理论框架。
text【匹配度】高 / 中 / 低（一句话理由） 【目标期刊】《经济学动态》 【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 | 31,837 | 17,277 | -46% | 1 | 1 | 0% | 3,428 | 3,442 | +0% | 0 | 0 | — |
case-02 | fail→pass | 123,639 | 19,568 | -84% | 1 | 1 | 0% | 3,376 | 3,400 | +1% | 0 | 0 | — |
case-03 | fail→pass | 30,637 | 18,511 | -40% | 1 | 1 | 0% | 3,300 | 3,400 | +3% | 0 | 0 | — |
case-08 | fail→pass | 24,718 | 21,495 | -13% | 1 | 1 | 0% | 3,178 | 3,579 | +13% | 0 | 0 | — |
case-09 | fail→fail | 25,496 | 23,194 | -9% | 1 | 1 | 0% | 3,137 | 3,773 | +20% | 0 | 0 | — |
case-10 | pass→pass | 21,689 | 29,884 | +38% | 1 | 1 | 0% | 2,735 | 3,770 | +38% | 0 | 0 | — |
case-04 | fail→pass | 24,546 | 27,285 | +11% | 1 | 1 | 0% | 2,670 | 4,070 | +52% | 0 | 0 | — |
case-05 | fail→pass | 26,406 | 23,182 | -12% | 1 | 1 | 0% | 2,735 | 3,895 | +42% | 0 | 0 | — |
case-06 | pass→pass | 21,557 | 15,223 | -29% | 1 | 1 | 0% | 2,204 | 3,035 | +38% | 0 | 0 | — |
case-07 | fail→pass | 25,616 | 22,927 | -10% | 1 | 1 | 0% | 2,842 | 4,108 | +45% | 0 | 0 | — |
case-11 | pass→pass | 20,063 | 7,843 | -61% | 1 | 1 | 0% | 1,965 | 2,224 | +13% | 0 | 0 | — |
case-12 | pass→pass | 28,027 | 31,771 | +13% | 1 | 1 | 0% | 3,138 | 4,633 | +48% | 0 | 0 | — |
case-13 | pass→pass | 18,242 | 25,192 | +38% | 1 | 1 | 0% | 2,454 | 3,886 | +58% | 0 | 0 | — |
case-14 | pass→pass | 27,725 | 34,267 | +24% | 1 | 1 | 0% | 3,095 | 3,787 | +22% | 0 | 0 | — |
case-15 | fail→fail | 28,948 | 24,897 | -14% | 1 | 1 | 0% | 2,950 | 4,350 | +47% | 0 | 0 | — |
case-16 | pass→pass | 23,918 | 21,276 | -11% | 1 | 1 | 0% | 2,384 | 3,470 | +46% | 0 | 0 | — |
case-17 | fail→pass | 30,172 | 24,979 | -17% | 1 | 1 | 0% | 3,535 | 3,991 | +13% | 0 | 0 | — |
case-18 | pass→pass | 22,224 | 25,146 | +13% | 1 | 1 | 0% | 2,695 | 4,801 | +78% | 0 | 0 | — |
case-19 | fail→fail | 8,811 | 9,090 | +3% | 1 | 1 | 0% | 501 | 1,941 | +287% | 0 | 0 | — |
case-20 | pass→pass | 66,178 | 67,200 | +2% | 1 | 1 | 0% | 8,225 | 9,590 | +17% | 0 | 0 | — |
case-21 | fail→pass | 29,091 | 20,755 | -29% | 1 | 1 | 0% | 2,940 | 3,431 | +17% | 0 | 0 | — |
case-22 | fail→fail | 29,340 | 19,456 | -34% | 1 | 1 | 0% | 3,147 | 4,168 | +32% | 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.