Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股个股复盘/交易复盘/个股走势回顾分析。当用户说"个股复盘"、"复盘"、"今天XX怎么回事"、"XX为什么涨"、"XX为什么跌"、"XX走势分析"、"stock review"、"盘后分析"、"XX今天异动"、"复盘分析"、"个股复盘分析"、"XX今天走势"、"XX涨跌原因"时触发。MUST USE when user asks about stock review, post-market analysis, why a stock went up/down today, or wants to review a specific stock's recent performance. 针对单只股票的当日或近期走势进行全面复盘，包括股价表现、量能变化、资金流向、消息面驱动因素。支持详细复盘风格（formal）和快速解读风格（brief）。
.claude/skills/aifinlab-a-share-stock-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-01 | ✓→✗ | ▼ Worse | -53% | 0% |
| case-06 | ✓→✗ | ▼ Worse | -61% | 0% |
| case-17 | ✓→✗ | ▼ Worse | -55% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 实时行情（价格/涨跌幅/换手率/量比/振幅等） python "$SCRIPTS/cn_stock_data.py" quote --code [CODE] # 近20日K线（趋势/支撑压力/均线） python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [20日前YYYYMMDD] # 资金流向（主力/散户净流入） python "$SCRIPTS/cn_stock_data.py" fund_flow --code [CODE] # 基本财务数据（市盈率/市净率/总市值等） python "$SCRIPTS/cn_stock_data.py" finance --code [CODE]
补充数据源：用 web 搜索获取当日公告、新闻、研报等消息面信息。
references/stock-review-template.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→fail | 19,985 | 19,969 | -0% | 1 | 1 | 0% | 3,060 | 1,443 | -53% | 0 | 0 | — |
case-02 | fail→fail | 11,810 | 20,829 | +76% | 1 | 1 | 0% | 1,467 | 3,239 | +121% | 0 | 0 | — |
case-03 | fail→fail | 25,186 | 11,355 | -55% | 1 | 1 | 0% | 3,256 | 1,372 | -58% | 0 | 0 | — |
case-04 | fail→fail | 65,586 | 10,354 | -84% | 1 | 1 | 0% | 3,642 | 1,385 | -62% | 0 | 0 | — |
case-05 | fail→fail | 7,420 | 30,481 | +311% | 1 | 1 | 0% | 874 | 2,999 | +243% | 0 | 0 | — |
case-06 | pass→fail | 23,894 | 62,955 | +163% | 1 | 1 | 0% | 2,835 | 1,095 | -61% | 0 | 0 | — |
case-17 | pass→fail | 20,175 | 13,739 | -32% | 1 | 1 | 0% | 2,906 | 1,303 | -55% | 0 | 0 | — |
case-07 | pass→pass | 15,807 | 18,147 | +15% | 1 | 1 | 0% | 1,958 | 3,183 | +63% | 0 | 0 | — |
case-08 | pass→fail | 15,853 | 12,133 | -23% | 1 | 1 | 0% | 2,201 | 1,017 | -54% | 0 | 0 | — |
case-09 | pass→pass | 23,312 | 21,321 | -9% | 1 | 1 | 0% | 3,140 | 3,778 | +20% | 0 | 0 | — |
case-10 | fail→fail | 36,459 | 21,679 | -41% | 1 | 1 | 0% | 2,966 | 3,535 | +19% | 0 | 0 | — |
case-11 | fail→fail | 31,888 | 10,947 | -66% | 1 | 1 | 0% | 2,964 | 1,231 | -58% | 0 | 0 | — |
case-12 | fail→fail | 27,700 | 19,275 | -30% | 1 | 1 | 0% | 2,510 | 3,502 | +40% | 0 | 0 | — |
case-13 | fail→pass | 14,250 | 10,818 | -24% | 1 | 1 | 0% | 2,688 | 1,887 | -30% | 0 | 0 | — |
case-14 | pass→fail | 13,428 | 5,342 | -60% | 1 | 1 | 0% | 1,954 | 1,390 | -29% | 0 | 0 | — |
case-15 | pass→pass | 17,103 | 7,527 | -56% | 1 | 1 | 0% | 2,674 | 1,771 | -34% | 0 | 0 | — |
case-16 | fail→pass | 16,364 | 18,132 | +11% | 1 | 1 | 0% | 2,269 | 2,962 | +31% | 0 | 0 | — |
case-18 | pass→pass | 17,762 | 9,970 | -44% | 1 | 1 | 0% | 2,270 | 2,243 | -1% | 0 | 0 | — |
case-19 | fail→fail | 12,549 | 10,154 | -19% | 1 | 1 | 0% | 1,777 | 1,189 | -33% | 0 | 0 | — |
case-20 | fail→fail | 26,584 | 28,748 | +8% | 1 | 1 | 0% | 5,154 | 5,867 | +14% | 0 | 0 | — |
case-21 | fail→fail | 25,871 | 33,456 | +29% | 1 | 1 | 0% | 3,979 | 5,296 | +33% | 0 | 0 | — |
case-22 | fail→fail | 35,725 | 44,198 | +24% | 1 | 1 | 0% | 6,860 | 7,794 | +14% | 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, and 12 counted toward the lift figure. The other 10 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -14 percentage points is the difference between those two pass rates over the 12 comparable cases. 11 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.