Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when targeting 《中国软科学》(China Soft Science — 科技部主管、中国软科学研究会与中国科学技术信息研究所主办的软科学权威月刊, 1986 年创刊) or deciding whether a Chinese innovation-strategy/policy manuscript fits this venue. Encodes the journal's fit, framing, low-similarity house style, official-submission re-check, and desk-reject heuristics.
.claude/skills/brycewang-stanford-china-soft-science/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-19 | ✗→✓ | ▲ Improved | 74% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 91% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 25% | 0% |
《中国软科学》由科学技术部主管，中国软科学研究会、中国科学技术信息研究所主办，1986 年创刊，月刊。它是软科学与科技政策领域的权威刊，面向创新战略、科技治理、产业政策、决策咨询与重大问题对策，强调对国家与区域层面热点、焦点问题的高水平研判。相对偏学术理论的《科学学研究》，本刊更看重问题的国家/战略落点与政策含义的厚度，但仍要求论据充分、方法规范、原创性高。
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。
china-rural-economy（《中国农村经济》） / china-rural-survey（《中国农村观察》） / chinese-journal-of-management（《管理学报》） / chinese-journal-of-management-science（《中国管理科学》）。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。更学术化的科学学/科技管理 → studies-in-science-of-science（《科学学研究》）/ science-of-science-and-management-of-st（《科学学与科学技术管理》）；偏决策模型与系统工程 → scientific-decision-making（《科学决策》）。
text【匹配度】高 / 中 / 低（一句话理由） 【目标期刊】《中国软科学》 【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/查重/摘要/参考文献/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-19 | fail→pass | 24,634 | 17,520 | -29% | 1 | 1 | 0% | 2,310 | 4,013 | +74% | 0 | 0 | — |
case-20 | fail→fail | 24,689 | 17,905 | -27% | 1 | 1 | 0% | 3,165 | 3,364 | +6% | 0 | 0 | — |
case-21 | pass→fail | 26,851 | 25,691 | -4% | 1 | 1 | 0% | 2,892 | 4,341 | +50% | 0 | 0 | — |
case-01 | fail→fail | 30,086 | 22,235 | -26% | 1 | 1 | 0% | 3,455 | 4,003 | +16% | 0 | 0 | — |
case-02 | fail→fail | 33,649 | 25,570 | -24% | 1 | 1 | 0% | 3,442 | 4,195 | +22% | 0 | 0 | — |
case-03 | fail→pass | 28,940 | 22,335 | -23% | 1 | 1 | 0% | 3,692 | 3,876 | +5% | 0 | 0 | — |
case-04 | pass→pass | 17,915 | 12,613 | -30% | 1 | 1 | 0% | 1,606 | 2,965 | +85% | 0 | 0 | — |
case-05 | pass→fail | 27,373 | 28,338 | +4% | 1 | 1 | 0% | 3,122 | 5,112 | +64% | 0 | 0 | — |
case-06 | pass→pass | 18,480 | 13,080 | -29% | 1 | 1 | 0% | 1,641 | 2,659 | +62% | 0 | 0 | — |
case-07 | fail→fail | 25,911 | 19,902 | -23% | 1 | 1 | 0% | 2,924 | 3,501 | +20% | 0 | 0 | — |
case-12 | fail→pass | 12,711 | 17,848 | +40% | 1 | 1 | 0% | 1,626 | 3,101 | +91% | 0 | 0 | — |
case-08 | fail→pass | 26,293 | 17,237 | -34% | 1 | 1 | 0% | 2,971 | 3,226 | +9% | 0 | 0 | — |
case-09 | fail→fail | 22,455 | 19,918 | -11% | 1 | 1 | 0% | 2,795 | 3,808 | +36% | 0 | 0 | — |
case-10 | pass→pass | 23,528 | 18,460 | -22% | 1 | 1 | 0% | 2,469 | 3,238 | +31% | 0 | 0 | — |
case-11 | fail→pass | 24,509 | 21,963 | -10% | 1 | 1 | 0% | 2,954 | 3,703 | +25% | 0 | 0 | — |
case-13 | fail→pass | 20,922 | 18,473 | -12% | 1 | 1 | 0% | 2,503 | 3,450 | +38% | 0 | 0 | — |
case-14 | fail→pass | 23,153 | 18,897 | -18% | 1 | 1 | 0% | 2,786 | 3,199 | +15% | 0 | 0 | — |
case-15 | fail→pass | 29,340 | 24,504 | -16% | 1 | 1 | 0% | 3,238 | 4,192 | +29% | 0 | 0 | — |
case-16 | fail→fail | 38,622 | 24,378 | -37% | 1 | 1 | 0% | 4,226 | 4,493 | +6% | 0 | 0 | — |
case-17 | fail→fail | 29,815 | 26,644 | -11% | 1 | 1 | 0% | 3,268 | 4,056 | +24% | 0 | 0 | — |
case-18 | fail→pass | 29,441 | 20,679 | -30% | 1 | 1 | 0% | 3,433 | 3,568 | +4% | 0 | 0 | — |
case-22 | pass→pass | 23,295 | 22,063 | -5% | 1 | 1 | 0% | 2,408 | 3,574 | +48% | 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 +32 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are 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.