Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →客户陪伴内容助手 - 行业热点版，专用于行业/主题热点解读的客户陪伴内容。 以下情况请主动触发此技能： - 某个行业/主题突然大涨/大跌，客户询问 - 重大产业政策发布，需要解读 - 行业出现黑天鹅事件，客户担忧 - 用户需要"行业点评""热点解读""主题分析"内容 - 用户准备行业轮动解读、主题投资机会分析 输出含行业解读、投资逻辑、风险提示的陪伴内容。 不要等用户明确说"行业陪伴"——只要涉及行业热点解读、主题分析、产业点评，就应主动启动此技能。
.claude/skills/aifinlab-client-engagement-industry-hotspot/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 57% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 70% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 31% | 0% |
核心定位：针对行业/主题热点事件，向客户提供专业、客观、及时的解读和陪伴内容。
适用场景：行业大涨大跌、产业政策发布、行业黑天鹅、主题轮动
| 热点类型 | 触发条件 | 内容基调 | |---------|---------|---------| | 行业大涨 | 单日>5% 或连续上涨 | 解读原因 + 持续性分析 + 不追高提醒 | | 行业大跌 | 单日>5% 或连续下跌 | 解读原因 + 影响评估 + 持仓建议 | | 政策利好 | 产业政策发布 | 政策解读 + 受益方向 + 理性看待 | | 政策利空 | 监管政策收紧 | 政策解读 + 影响范围 + 长期逻辑 | | 黑天鹅事件 | 行业突发负面 | 事件说明 + 影响评估 + 区分个股 | | 主题轮动 | 资金切换明显 | 轮动逻辑 + 持续性判断 + 配置建议 |
【行业热点解读 | XX 行业】
📈 行情说明
- 今日 XX 行业大涨 X%，领涨市场
- 板块内多只个股涨停
🔍 上涨原因
1. [原因 1：如政策利好]
2. [原因 2：如业绩超预期]
3. [原因 3：如资金轮动]
💡 持续性分析
- 短期：[资金情绪/技术面]
- 中期：[基本面/估值]
- 长期：[行业逻辑]
⚠️ 风险提示
- 短期涨幅较大，注意波动
- 不建议追高加仓
📌 建议
- 持仓的继续持有
- 想布局的等待回调
- 关注 [细分方向]【行业波动解读 | XX 行业】
📉 行情说明
- 今日 XX 行业大跌 X%，领跌市场
- 主要受 [原因] 影响
🔍 下跌原因
1. [直接原因]
2. [深层原因]
3. [情绪因素]
💡 影响评估
- 短期影响：[评估]
- 长期逻辑：[是否变化]
- 估值位置：[历史分位]
📌 持仓建议
- 检视持仓逻辑是否变化
- 优质资产值得坚守
- 仓位过高的可适当调整
⚠️ 风险提示
- 行业波动可能持续
- 注意分散配置【政策速递 | XX 政策】
📋 政策要点
- [政策名称] 发布
- 核心内容：[1-2 句概括]
- 实施时间：[时间]
🎯 受益方向
1. [细分领域 1]
2. [细分领域 2]
3. [细分领域 3]
💡 投资逻辑
- 政策背景：[为什么出台]
- 行业影响：[需求/供给/竞争格局]
- 长期空间：[市场规模]
⚠️ 理性看待
- 政策落地需要时间
- 避免追高，关注估值
- 优选 [筛选标准]【突发事件解读 | XX 事件】
🚨 事件说明
- [事件简述]
- 影响范围：[行业/公司]
🔍 影响评估
- 直接影响：[评估]
- 间接影响：[评估]
- 情绪影响：[评估]
📌 应对建议
- 区分行业 vs 个股
- 检视持仓受影响程度
- 避免恐慌性操作
💡 长期视角
- 事件性质：[一次性/持续性]
- 行业逻辑：[是否变化]
- 历史参考：[类似案例]关注点：
关注点：
关注点：
关注点：
关注点：
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 19,738 | 20,218 | +2% | 1 | 1 | 0% | 2,773 | 3,880 | +40% | 0 | 0 | — |
case-02 | fail→fail | 19,888 | 17,435 | -12% | 1 | 1 | 0% | 2,748 | 4,070 | +48% | 0 | 0 | — |
case-03 | fail→fail | 17,624 | 37,456 | +113% | 1 | 1 | 0% | 2,565 | 4,113 | +60% | 0 | 0 | — |
case-04 | fail→fail | 19,658 | 20,213 | +3% | 1 | 1 | 0% | 2,786 | 4,503 | +62% | 0 | 0 | — |
case-05 | pass→pass | 18,286 | 15,726 | -14% | 1 | 1 | 0% | 2,637 | 3,839 | +46% | 0 | 0 | — |
case-06 | fail→pass | 21,992 | 17,574 | -20% | 1 | 1 | 0% | 2,935 | 4,021 | +37% | 0 | 0 | — |
case-07 | fail→fail | 20,616 | 16,492 | -20% | 1 | 1 | 0% | 2,872 | 3,990 | +39% | 0 | 0 | — |
case-08 | pass→pass | 21,127 | 14,215 | -33% | 1 | 1 | 0% | 3,010 | 3,556 | +18% | 0 | 0 | — |
case-09 | fail→fail | 25,327 | 14,480 | -43% | 1 | 1 | 0% | 3,419 | 3,645 | +7% | 0 | 0 | — |
case-10 | fail→fail | 19,856 | 15,513 | -22% | 1 | 1 | 0% | 2,773 | 3,919 | +41% | 0 | 0 | — |
case-11 | pass→pass | 18,512 | 14,616 | -21% | 1 | 1 | 0% | 2,496 | 3,636 | +46% | 0 | 0 | — |
case-12 | pass→pass | 16,827 | 13,502 | -20% | 1 | 1 | 0% | 2,258 | 3,520 | +56% | 0 | 0 | — |
case-13 | fail→fail | 22,189 | 16,541 | -25% | 1 | 1 | 0% | 3,103 | 3,916 | +26% | 0 | 0 | — |
case-14 | fail→fail | 16,445 | 14,851 | -10% | 1 | 1 | 0% | 2,261 | 3,614 | +60% | 0 | 0 | — |
case-15 | fail→fail | 19,072 | 18,184 | -5% | 1 | 1 | 0% | 2,490 | 4,022 | +62% | 0 | 0 | — |
case-16 | pass→pass | 18,362 | 13,999 | -24% | 1 | 1 | 0% | 2,314 | 3,592 | +55% | 0 | 0 | — |
case-17 | fail→pass | 19,150 | 18,406 | -4% | 1 | 1 | 0% | 2,681 | 4,207 | +57% | 0 | 0 | — |
case-18 | fail→pass | 18,092 | 34,015 | +88% | 1 | 1 | 0% | 2,446 | 4,167 | +70% | 0 | 0 | — |
case-19 | fail→pass | 19,887 | 16,000 | -20% | 1 | 1 | 0% | 2,664 | 3,750 | +41% | 0 | 0 | — |
case-20 | fail→pass | 19,968 | 16,732 | -16% | 1 | 1 | 0% | 3,027 | 3,961 | +31% | 0 | 0 | — |
case-21 | fail→fail | 18,880 | 16,011 | -15% | 1 | 1 | 0% | 3,002 | 4,101 | +37% | 0 | 0 | — |
case-22 | fail→fail | 27,740 | 26,875 | -3% | 1 | 1 | 0% | 3,577 | 4,976 | +39% | 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 +23 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.