Install any skill in seconds. Free to start, no credit card required.
Get Started Free →客户陪伴内容助手 - 市场波动版,专用于市场大幅波动时向客户发送陪伴内容。 以下情况请主动触发此技能: - 市场单日涨跌>2%,需要向客户发送安抚/解读内容 - 连续多日下跌,客户焦虑情绪上升 - 市场快速上涨,需要提醒客户理性看待 - 用户需要"市场点评""波动解读""安抚文案""持有信心维护"内容 - 用户准备群发内容、朋友圈文案、一对一沟通话术 输出含市场解读、情绪安抚、长期理念传递的陪伴内容。 不要等用户明确说"市场陪伴"——只要涉及市场波动时的客户安抚、解读、陪伴,就应主动启动此技能。
.claude/skills/aifinlab-client-engagement-market-volatility/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-12 | ✓→✗ | ▼ Worse | 66% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 54% | 0% |
核心定位:在市场大幅波动时,向客户传递理性、专业、有温度的陪伴内容,稳定客户情绪,传递长期投资理念。
适用场景:市场单日涨跌>2%、连续波动、极端行情
| 波动类型 | 触发条件 | 内容基调 | |---------|---------|---------| | 单日大跌 | 单日跌幅>2% | 安抚情绪 + 解读原因 + 长期理念 | | 连续下跌 | 3 日累计>5% | 深度解读 + 历史参考 + 配置建议 | | 单日大涨 | 单日涨幅>2% | 理性看待 + 风险提示 + 不追高 | | 快速上涨 | 短期累计>10% | 提醒止盈 + 估值分析 + 长期视角 | | 震荡行情 | 持续横盘震荡 | 保持耐心 + 定投建议 + 时间价值 |
【市场波动提醒】
今日市场大幅调整,沪深 300 下跌 X%,主要受 [原因] 影响。
理解大家的焦虑,市场波动是常态,历史上类似情况多次出现。
长期来看,[核心理念:如"经济基本面未变"/"估值已处低位"]。
建议:
1. 保持冷静,不要盲目操作
2. 检视持仓,优质资产值得坚守
3. 有闲钱可考虑分批布局
市场波动期,我们与您一起坚守。【致投资者的一封信】
最近市场连续调整,很多朋友感到焦虑,这很正常。
📊 市场情况
- 近 X 日累计下跌 X%
- 主要受 [因素 1]、[因素 2] 影响
- 当前估值处于历史 [分位]
📚 历史参考
- 20XX 年类似情况:下跌 X% 后,X 个月内修复
- 20XX 年:[另一个历史案例]
- 结论:波动是暂时的,价值终会回归
💡 我们的建议
1. 检视持仓逻辑是否变化
2. 避免情绪化操作
3. 闲钱可分批布局优质资产
4. 做好资产配置,控制仓位
🌟 长期理念
投资是认知的变现,波动是收益的来源。
在别人恐惧时保持理性,时间会奖励耐心的人。
市场低迷期,我们与您同行。【市场快速上涨提醒】
今日市场大幅上涨 X%,恭喜持仓的朋友!
理性看待:
- 上涨受 [原因] 推动
- 短期涨幅较大,注意波动风险
- 不建议追高加仓
建议:
1. 持仓的继续持有,享受上涨
2. 想加仓的等待回调
3. 浮盈较多的可考虑部分止盈
投资是长跑,不为一时涨跌所动。【开场】
"XX 总,今天市场波动比较大,给您发个消息。"
【共情】
"知道您可能有些担心,这很正常,市场确实跌了不少。"
【解读】
"这次下跌主要是 [原因],从基本面看 [分析]。"
【建议】
"我的建议是 [具体建议],您看怎么样?"
【结尾】
"有任何问题随时联系我,我们一起应对。"内容要点:
内容要点:
内容要点:
内容要点:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,021 | 43,386 | +171% | 1 | 1 | 0% | 2,229 | 3,362 | +51% | 0 | 0 | — |
case-02 | pass→pass | 17,388 | 16,631 | -4% | 1 | 1 | 0% | 2,396 | 3,682 | +54% | 0 | 0 | — |
case-03 | pass→pass | 21,101 | 22,743 | +8% | 1 | 1 | 0% | 3,172 | 4,675 | +47% | 0 | 0 | — |
case-04 | pass→pass | 23,690 | 23,829 | +1% | 1 | 1 | 0% | 4,153 | 5,604 | +35% | 0 | 0 | — |
case-05 | pass→pass | 23,623 | 20,927 | -11% | 1 | 1 | 0% | 3,502 | 4,375 | +25% | 0 | 0 | — |
case-06 | fail→fail | 20,552 | 17,632 | -14% | 1 | 1 | 0% | 2,711 | 3,799 | +40% | 0 | 0 | — |
case-07 | pass→pass | 22,903 | 18,549 | -19% | 1 | 1 | 0% | 3,111 | 3,745 | +20% | 0 | 0 | — |
case-08 | pass→pass | 14,844 | 14,334 | -3% | 1 | 1 | 0% | 2,198 | 3,498 | +59% | 0 | 0 | — |
case-09 | pass→pass | 14,214 | 14,570 | +3% | 1 | 1 | 0% | 2,172 | 3,536 | +63% | 0 | 0 | — |
case-10 | pass→pass | 15,160 | 16,236 | +7% | 1 | 1 | 0% | 2,246 | 3,752 | +67% | 0 | 0 | — |
case-11 | fail→pass | 20,064 | 20,759 | +3% | 1 | 1 | 0% | 2,857 | 4,450 | +56% | 0 | 0 | — |
case-12 | pass→fail | 16,724 | 17,968 | +7% | 1 | 1 | 0% | 2,334 | 3,869 | +66% | 0 | 0 | — |
case-13 | fail→fail | 22,079 | 20,106 | -9% | 1 | 1 | 0% | 2,865 | 4,307 | +50% | 0 | 0 | — |
case-14 | pass→pass | 22,632 | 22,550 | -0% | 1 | 1 | 0% | 3,025 | 4,460 | +47% | 0 | 0 | — |
case-15 | fail→pass | 16,281 | 12,718 | -22% | 1 | 1 | 0% | 2,450 | 3,323 | +36% | 0 | 0 | — |
case-16 | pass→pass | 16,069 | 8,849 | -45% | 1 | 1 | 0% | 2,244 | 2,839 | +27% | 0 | 0 | — |
case-17 | fail→fail | 19,579 | 18,204 | -7% | 1 | 1 | 0% | 2,738 | 3,842 | +40% | 0 | 0 | — |
case-18 | pass→pass | 14,373 | 11,981 | -17% | 1 | 1 | 0% | 1,949 | 3,058 | +57% | 0 | 0 | — |
case-19 | fail→pass | 20,579 | 14,290 | -31% | 1 | 1 | 0% | 2,871 | 3,484 | +21% | 0 | 0 | — |
case-20 | pass→pass | 16,816 | 18,116 | +8% | 1 | 1 | 0% | 2,265 | 3,764 | +66% | 0 | 0 | — |
case-21 | pass→pass | 12,968 | 11,517 | -11% | 1 | 1 | 0% | 2,006 | 3,268 | +63% | 0 | 0 | — |
case-22 | pass→pass | 10,362 | 13,511 | +30% | 1 | 1 | 0% | 1,620 | 3,496 | +116% | 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 +9 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.