Install any skill in seconds. Free to start, no credit card required.
Get Started Free →定位图 + 功能矩阵 + 价格对比 + 机会窗口, 把竞品资料转成产品决策报告
.claude/skills/nexu-io-competitive-teardown/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 93% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 377% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 888% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 454% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 465% | 0% |
【模板: 竞品拆解 / Competitive Teardown】 【意图】这不是文章、不是 PRD、不是 pitch deck。目标是把多个竞品的杂乱资料转成一份可决策的产品战略报告, 帮团队回答: "我们和它们到底差在哪里, 下一步该怎么打?"
【适合输入】
【必须输出的结构】
【设计要求】
【可选风格模板 — 参考 assets/】 根据用户内容选择最贴合的一种, 不要三种混用:
assets/war-room-grid.html: 默认风格。浅色战情室 / 咨询报告, 适合产品团队、PM、普通商业读者。assets/radar-map.html: 深色雷达图 / market intelligence console, 适合安全、AI、开发者工具、平台型竞品。assets/analyst-dossier.html: 纸质分析档案 / investment research dossier, 适合投研、行业分析、正式战略备忘。如果用户没有指定风格, 优先使用 war-room-grid; 如果输入强调市场格局、技术雷达、攻防态势, 使用 radar-map; 如果输入像研究笔记、投资备忘或行业报告, 使用 analyst-dossier。
【内容真实性】
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-14 | pass→fail | 11,935 | 27,337 | +129% | 1 | 1 | 0% | 2,133 | 6,908 | +224% | 0 | 0 | — |
case-01 | pass→pass | 21,613 | 28,161 | +30% | 1 | 1 | 0% | 3,585 | 6,922 | +93% | 0 | 0 | — |
case-15 | fail→fail | 12,225 | 26,574 | +117% | 1 | 1 | 0% | 1,923 | 6,907 | +259% | 0 | 0 | — |
case-02 | fail→fail | 5,365 | 28,441 | +430% | 1 | 1 | 0% | 852 | 6,915 | +712% | 0 | 0 | — |
case-03 | pass→fail | 16,010 | 26,023 | +63% | 1 | 1 | 0% | 2,625 | 6,919 | +164% | 0 | 0 | — |
case-04 | fail→fail | 11,973 | 24,612 | +106% | 1 | 1 | 0% | 2,091 | 5,326 | +155% | 0 | 0 | — |
case-05 | fail→pass | 14,959 | 25,536 | +71% | 1 | 1 | 0% | 2,740 | 5,277 | +93% | 0 | 0 | — |
case-16 | pass→pass | 21,739 | 28,561 | +31% | 1 | 1 | 0% | 3,747 | 6,919 | +85% | 0 | 0 | — |
case-06 | fail→fail | 7,934 | 25,015 | +215% | 1 | 1 | 0% | 1,406 | 6,911 | +392% | 0 | 0 | — |
case-07 | fail→pass | 6,486 | 26,218 | +304% | 1 | 1 | 0% | 1,073 | 5,118 | +377% | 0 | 0 | — |
case-08 | fail→pass | 3,792 | 33,285 | +778% | 1 | 1 | 0% | 668 | 6,598 | +888% | 0 | 0 | — |
case-09 | fail→fail | 20,425 | 26,405 | +29% | 1 | 1 | 0% | 3,397 | 5,048 | +49% | 0 | 0 | — |
case-10 | fail→fail | 4,887 | 24,336 | +398% | 1 | 1 | 0% | 824 | 4,795 | +482% | 0 | 0 | — |
case-11 | fail→fail | 13,483 | 26,173 | +94% | 1 | 1 | 0% | 2,245 | 6,909 | +208% | 0 | 0 | — |
case-12 | fail→pass | 7,502 | 35,248 | +370% | 1 | 1 | 0% | 1,248 | 6,908 | +454% | 0 | 0 | — |
case-13 | fail→pass | 7,478 | 27,845 | +272% | 1 | 1 | 0% | 1,222 | 6,910 | +465% | 0 | 0 | — |
case-17 | fail→pass | 16,442 | 27,559 | +68% | 1 | 1 | 0% | 1,747 | 4,947 | +183% | 0 | 0 | — |
case-18 | fail→pass | 7,811 | 23,880 | +206% | 1 | 1 | 0% | 1,181 | 4,409 | +273% | 0 | 0 | — |
case-19 | fail→pass | 6,684 | 26,990 | +304% | 1 | 1 | 0% | 969 | 6,908 | +613% | 0 | 0 | — |
case-20 | pass→fail | 19,325 | 32,856 | +70% | 1 | 1 | 0% | 3,325 | 6,916 | +108% | 0 | 0 | — |
case-21 | pass→pass | 11,249 | 13,373 | +19% | 1 | 1 | 0% | 2,132 | 3,033 | +42% | 0 | 0 | — |
case-22 | pass→fail | 15,938 | 24,917 | +56% | 1 | 1 | 0% | 2,642 | 6,913 | +162% | 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 +18 percentage points is the difference between those two pass rates over the 22 comparable cases. 4 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.