Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股周度复盘/每周总结。当用户说"周复盘"、"本周总结"、"这周行情回顾"、"weekly review"、"周报"、"这周市场怎么样"、"一周回顾"、"周末复盘"时触发。基于 cn-stock-data 获取本周大盘走势、板块涨跌、资金流向、重要事件，生成结构化周度复盘报告。支持券商周报风格（formal）和个人复盘笔记风格（brief）。
.claude/skills/aifinlab-a-share-weekly-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-09 | ✓→✗ | ▼ Worse | 15% | 0% |
| case-10 | ✓→✗ | ▼ Worse | -45% | 0% |
| case-11 | ✓→✗ | ▼ Worse | -11% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 大盘指数本周K线（将 MONDAY 替换为本周一日期 YYYYMMDD） python "$SCRIPTS/cn_stock_data.py" kline --code SH000001,SZ399001,SZ399006 --freq daily --start MONDAY # 实时行情（最新价/涨跌幅） python "$SCRIPTS/cn_stock_data.py" quote --code SH000001,SZ399001,SZ399006 # 主力资金流向 python "$SCRIPTS/cn_stock_data.py" fund_flow --code SH000001
补充数据通过 web 搜索获取：板块涨跌排名、北向资金周净流入、融资余额、本周重要事件/政策。
默认 formal；用户说"简单总结/简要/brief"时用 brief。
| 维度 | formal | brief | |------|--------|-------| | 篇幅 | 2-3 页 | 半页 | | 板块分析 | 完整行业涨跌榜 | Top 3 涨跌 | | 个股 | 本周牛股/熊股各5只 | 仅提亮点 | | 下周展望 | 多维度分析 | 1-2 句关键判断 | | 免责声明 | 需要 | 不需要 |
references/weekly-review-template.md> 以上内容仅供参考，不构成投资建议。市场有风险，投资需谨慎。
python# 调用 skill result = run_skill({ "param1": "value1", "param2": "value2" })
bashpython scripts/run_skill.py --input data.json
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 25,842 | 17,115 | -34% | 1 | 1 | 0% | 3,414 | 1,258 | -63% | 0 | 0 | — |
case-02 | fail→fail | 12,667 | 8,470 | -33% | 1 | 1 | 0% | 1,818 | 1,299 | -29% | 0 | 0 | — |
case-03 | fail→fail | 22,521 | 8,220 | -64% | 1 | 1 | 0% | 3,316 | 1,235 | -63% | 0 | 0 | — |
case-04 | pass→pass | 31,678 | 5,864 | -81% | 1 | 1 | 0% | 3,138 | 1,696 | -46% | 0 | 0 | — |
case-05 | fail→pass | 16,669 | 3,230 | -81% | 1 | 1 | 0% | 2,738 | 1,411 | -48% | 0 | 0 | — |
case-06 | fail→fail | 16,503 | 8,543 | -48% | 1 | 1 | 0% | 2,639 | 2,193 | -17% | 0 | 0 | — |
case-07 | pass→pass | 9,511 | 6,216 | -35% | 1 | 1 | 0% | 1,485 | 1,771 | +19% | 0 | 0 | — |
case-08 | pass→pass | 14,096 | 4,039 | -71% | 1 | 1 | 0% | 2,180 | 1,364 | -37% | 0 | 0 | — |
case-09 | pass→fail | 8,282 | 9,049 | +9% | 1 | 1 | 0% | 1,240 | 1,426 | +15% | 0 | 0 | — |
case-10 | pass→fail | 14,084 | 3,664 | -74% | 1 | 1 | 0% | 2,467 | 1,350 | -45% | 0 | 0 | — |
case-11 | pass→fail | 10,392 | 9,439 | -9% | 1 | 1 | 0% | 1,477 | 1,309 | -11% | 0 | 0 | — |
case-12 | fail→fail | 22,510 | 7,136 | -68% | 1 | 1 | 0% | 3,698 | 1,330 | -64% | 0 | 0 | — |
case-13 | fail→fail | 11,859 | 7,975 | -33% | 1 | 1 | 0% | 1,535 | 1,153 | -25% | 0 | 0 | — |
case-14 | fail→fail | 15,811 | 12,850 | -19% | 1 | 1 | 0% | 2,521 | 2,929 | +16% | 0 | 0 | — |
case-15 | fail→fail | 16,202 | 5,825 | -64% | 1 | 1 | 0% | 2,192 | 1,538 | -30% | 0 | 0 | — |
case-16 | pass→pass | 11,188 | 5,122 | -54% | 1 | 1 | 0% | 2,151 | 1,608 | -25% | 0 | 0 | — |
case-17 | fail→pass | 9,728 | 4,694 | -52% | 1 | 1 | 0% | 1,648 | 1,604 | -3% | 0 | 0 | — |
case-18 | pass→pass | 8,741 | 5,534 | -37% | 1 | 1 | 0% | 1,405 | 1,532 | +9% | 0 | 0 | — |
case-19 | fail→fail | 10,860 | 2,695 | -75% | 1 | 1 | 0% | 1,524 | 1,242 | -19% | 0 | 0 | — |
case-20 | pass→pass | 19,060 | 21,328 | +12% | 1 | 1 | 0% | 2,872 | 3,718 | +29% | 0 | 0 | — |
case-21 | fail→fail | 17,212 | 25,432 | +48% | 1 | 1 | 0% | 2,696 | 4,702 | +74% | 0 | 0 | — |
case-22 | fail→fail | 16,021 | 17,434 | +9% | 1 | 1 | 0% | 2,442 | 1,234 | -49% | 0 | 0 | — |
case-23 | fail→fail | 28,161 | 34,503 | +23% | 1 | 1 | 0% | 3,687 | 5,366 | +46% | 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. 23 cases were attempted, and 15 counted toward the lift figure. The other 8 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 -4 percentage points is the difference between those two pass rates over the 15 comparable cases. 5 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.