Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股主题投资/赛道分析。当用户说"主题投资"、"赛道"、"风口"、"thematic"、"XX赛道怎么样"、"AI投资机会"、"新能源赛道"、"半导体赛道"、"医药赛道"、"国产替代"、"数字经济"、"新质生产力"时触发。深度分析特定投资主题/赛道的产业逻辑、市场空间、竞争格局、估值水平和核心标的，评估主题的持续性和投资时机。支持研报风格（formal）和快速分析风格（brief）。
.claude/skills/aifinlab-a-share-thematic-invest/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 146% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 92% | 0% |
| case-20 | ✓→✗ | ▼ Worse | -20% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 44% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 概念板块列表（akshare，获取主题相关概念板块） python -c "import akshare as ak; df=ak.stock_board_concept_name_em(); print(df.to_json(orient='records', force_ascii=False))" # 概念板块成分股（akshare，需指定板块名称） python -c "import akshare as ak; df=ak.stock_board_concept_cons_em(symbol='人工智能'); print(df.to_json(orient='records', force_ascii=False))" # 核心标的实时行情 python "$SCRIPTS/cn_stock_data.py" quote --code [CODE1],[CODE2],... # 核心标的K线（判断趋势） python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [日期] # 核心标的财务数据（估值/业绩弹性） python "$SCRIPTS/cn_stock_data.py" finance --code [CODE1],[CODE2],... # 核心标的资金流向 python "$SCRIPTS/cn_stock_data.py" fund_flow --code [CODE]
补充：通过 web 搜索获取产业政策、市场空间/TAM数据、渗透率、技术进展、研报观点。
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 | 36,510 | 27,291 | -25% | 1 | 1 | 0% | 5,559 | 1,787 | -68% | 0 | 0 | — |
case-02 | fail→fail | 21,335 | 11,229 | -47% | 1 | 1 | 0% | 3,179 | 1,518 | -52% | 0 | 0 | — |
case-03 | fail→fail | 33,596 | 14,908 | -56% | 1 | 1 | 0% | 5,247 | 2,173 | -59% | 0 | 0 | — |
case-04 | pass→pass | 18,579 | 19,434 | +5% | 1 | 1 | 0% | 2,819 | 4,073 | +44% | 0 | 0 | — |
case-05 | pass→pass | 11,552 | 11,282 | -2% | 1 | 1 | 0% | 1,889 | 2,792 | +48% | 0 | 0 | — |
case-16 | pass→pass | 21,429 | 20,548 | -4% | 1 | 1 | 0% | 3,098 | 4,320 | +39% | 0 | 0 | — |
case-06 | fail→pass | 13,370 | 15,926 | +19% | 1 | 1 | 0% | 2,100 | 2,673 | +27% | 0 | 0 | — |
case-07 | pass→pass | 22,019 | 17,862 | -19% | 1 | 1 | 0% | 2,970 | 3,880 | +31% | 0 | 0 | — |
case-08 | fail→pass | 16,329 | 28,274 | +73% | 1 | 1 | 0% | 2,163 | 5,326 | +146% | 0 | 0 | — |
case-09 | pass→pass | 13,419 | 13,245 | -1% | 1 | 1 | 0% | 2,072 | 2,985 | +44% | 0 | 0 | — |
case-10 | pass→pass | 20,452 | 20,379 | -0% | 1 | 1 | 0% | 2,428 | 3,817 | +57% | 0 | 0 | — |
case-11 | fail→pass | 11,106 | 15,473 | +39% | 1 | 1 | 0% | 1,736 | 3,336 | +92% | 0 | 0 | — |
case-12 | pass→pass | 14,520 | 11,435 | -21% | 1 | 1 | 0% | 2,192 | 2,691 | +23% | 0 | 0 | — |
case-13 | pass→pass | 13,331 | 5,444 | -59% | 1 | 1 | 0% | 2,102 | 1,868 | -11% | 0 | 0 | — |
case-14 | pass→pass | 10,510 | 9,256 | -12% | 1 | 1 | 0% | 1,496 | 2,420 | +62% | 0 | 0 | — |
case-15 | pass→pass | 19,119 | 21,972 | +15% | 1 | 1 | 0% | 2,915 | 4,411 | +51% | 0 | 0 | — |
case-17 | pass→pass | 13,440 | 15,164 | +13% | 1 | 1 | 0% | 1,972 | 2,871 | +46% | 0 | 0 | — |
case-18 | pass→pass | 8,111 | 7,197 | -11% | 1 | 1 | 0% | 1,318 | 2,311 | +75% | 0 | 0 | — |
case-19 | pass→pass | 12,894 | 9,772 | -24% | 1 | 1 | 0% | 1,848 | 2,616 | +42% | 0 | 0 | — |
case-20 | pass→fail | 13,883 | 8,757 | -37% | 1 | 1 | 0% | 2,216 | 1,762 | -20% | 0 | 0 | — |
case-21 | pass→pass | 20,942 | 25,876 | +24% | 1 | 1 | 0% | 4,116 | 5,134 | +25% | 0 | 0 | — |
case-22 | pass→pass | 12,057 | 13,274 | +10% | 1 | 1 | 0% | 2,161 | 3,566 | +65% | 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 18 counted toward the lift figure. The other 4 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 +9 percentage points is the difference between those two pass rates over the 18 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.