Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when targeting 《经济学家》(Economist — 西南财经大学/四川社科学术基金会主办的理论经济学月刊, 1989 年创刊, CSSCI) or deciding whether a Chinese econ/policy manuscript fits this venue. Encodes the journal's fit, framing, abstract house style, official-submission re-check, and desk-reject heuristics.
.claude/skills/brycewang-stanford-economist-cn/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 32% | 0% |
《经济学家》由西南财经大学、四川社会科学学术基金会主办，是理论经济学月刊，1989 年创刊（CN 51-1312/F，ISSN 1003-5656），CSSCI 来源期刊。重视经济理论、中国式现代化、中国经济重大现实问题与政策研究——宏观调控、财政金融、产业政策、收入分配、共同富裕与高质量发展。强调理论深度与原创性。
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。
economic-science（《经济科学》） / economic-theory-and-business-management（《经济理论与经济管理》） / finance-and-economics（《财经科学》） / finance-and-trade-economics（《财贸经济》）。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。财政金融更专业转 finance-and-economics/public-finance-research。
text【匹配度】高 / 中 / 低（一句话理由） 【目标期刊】《经济学家》 【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 | 24,067 | 19,070 | -21% | 1 | 1 | 0% | 2,997 | 3,428 | +14% | 0 | 0 | — |
case-02 | fail→fail | 27,271 | 19,699 | -28% | 1 | 1 | 0% | 2,846 | 3,248 | +14% | 0 | 0 | — |
case-03 | fail→pass | 32,244 | 18,006 | -44% | 1 | 1 | 0% | 3,559 | 3,287 | -8% | 0 | 0 | — |
case-11 | fail→pass | 42,060 | 28,020 | -33% | 1 | 1 | 0% | 3,623 | 3,942 | +9% | 0 | 0 | — |
case-12 | fail→fail | 31,215 | 25,603 | -18% | 1 | 1 | 0% | 3,140 | 4,146 | +32% | 0 | 0 | — |
case-10 | pass→pass | 19,530 | 16,007 | -18% | 1 | 1 | 0% | 2,209 | 2,887 | +31% | 0 | 0 | — |
case-09 | pass→pass | 27,911 | 23,227 | -17% | 1 | 1 | 0% | 2,720 | 3,837 | +41% | 0 | 0 | — |
case-04 | pass→pass | 33,921 | 40,657 | +20% | 1 | 1 | 0% | 3,974 | 6,140 | +55% | 0 | 0 | — |
case-05 | pass→fail | 16,592 | 19,188 | +16% | 1 | 1 | 0% | 1,513 | 3,924 | +159% | 0 | 0 | — |
case-06 | pass→pass | 18,491 | 16,927 | -8% | 1 | 1 | 0% | 1,972 | 2,889 | +47% | 0 | 0 | — |
case-07 | fail→pass | 22,139 | 19,232 | -13% | 1 | 1 | 0% | 2,619 | 3,545 | +35% | 0 | 0 | — |
case-08 | fail→fail | 24,109 | 15,488 | -36% | 1 | 1 | 0% | 2,795 | 3,319 | +19% | 0 | 0 | — |
case-13 | pass→pass | 32,460 | 29,313 | -10% | 1 | 1 | 0% | 3,398 | 4,471 | +32% | 0 | 0 | — |
case-14 | fail→pass | 21,728 | 17,396 | -20% | 1 | 1 | 0% | 2,139 | 2,826 | +32% | 0 | 0 | — |
case-15 | fail→fail | 27,416 | 28,584 | +4% | 1 | 1 | 0% | 2,939 | 4,101 | +40% | 0 | 0 | — |
case-16 | fail→pass | 25,986 | 19,918 | -23% | 1 | 1 | 0% | 2,545 | 3,617 | +42% | 0 | 0 | — |
case-17 | fail→pass | 31,557 | 19,317 | -39% | 1 | 1 | 0% | 3,167 | 3,540 | +12% | 0 | 0 | — |
case-18 | fail→pass | 20,566 | 20,292 | -1% | 1 | 1 | 0% | 2,481 | 3,322 | +34% | 0 | 0 | — |
case-19 | fail→fail | 20,761 | 11,006 | -47% | 1 | 1 | 0% | 2,142 | 2,111 | -1% | 0 | 0 | — |
case-20 | fail→fail | 11,892 | 8,642 | -27% | 1 | 1 | 0% | 1,480 | 1,792 | +21% | 0 | 0 | — |
case-21 | fail→fail | 27,788 | 23,368 | -16% | 1 | 1 | 0% | 2,700 | 4,063 | +50% | 0 | 0 | — |
case-22 | fail→pass | 21,996 | 14,076 | -36% | 1 | 1 | 0% | 3,066 | 2,870 | -6% | 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 +36 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.