Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股主力资金/大单追踪/资金流向分析。当用户说"主力资金"、"大单"、"资金流向"、"主力在干嘛"、"资金净流入"、"主力买入"、"大资金"、"XX的资金流向"、"主力动向"、"smart money"、"主力资金流入流出"、"资金流向分析"、"大单净买入"、"超大单"、"主力净流入"时触发。MUST USE when user asks about main capital flow, large order tracking, money flow analysis, net inflow/outflow of institutional money, or smart money movement for individual stocks or the overall market. 基于 cn-stock-data 资金流向数据（超大单/大单/中单/小单），分析个股或市场的主力资金行为。支持机构级分析（formal）和快速判断（brief）。与 a-share-northbound 的区别：本 skill 聚焦场内主力资金（大单/超大单），northbound 聚焦北向外资。
.claude/skills/aifinlab-a-share-money-flow/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-04 | ✓→✗ | ▼ Worse | -68% | 0% |
| case-05 | ✓→✗ | ▼ Worse | -55% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 6% | 0% |
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" # 个股资金流向（含超大单/大单/中单/小单） python "$SCRIPTS/cn_stock_data.py" fund_flow --code [CODE] # 个股行情 python "$SCRIPTS/cn_stock_data.py" quote --code [CODE] # K线（叠加资金流分析） python "$SCRIPTS/cn_stock_data.py" kline --code [CODE] --freq daily --start [日期]
补充：web 搜索获取全市场资金流向排名（净流入/净流出 Top 20）。
| 维度 | formal | brief | |------|--------|-------| | 输出 | 完整资金分析报告 | 资金方向一句话 | | 分层 | 4层（超大/大/中/小） | 仅主力（超大+大） | | 趋势 | 近5/10/20日趋势 | 近3日方向 | | 排名 | 全市场资金流Top20 | 仅标的数据 |
默认 brief。用户要求"详细"、"报告"、"完整分析"时用 formal。
references/money-flow-guide.md 了解数据原理和信号矩阵在 formal 输出末尾附加：资金流向数据基于成交量估算，仅供参考，不构成投资建议。
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-08 | fail→fail | 28,248 | 9,798 | -65% | 1 | 1 | 0% | 3,618 | 1,285 | -64% | 0 | 0 | — |
case-01 | fail→fail | 31,081 | 12,017 | -61% | 1 | 1 | 0% | 4,548 | 1,478 | -68% | 0 | 0 | — |
case-02 | fail→fail | 14,319 | 36,314 | +154% | 1 | 1 | 0% | 1,927 | 1,315 | -32% | 0 | 0 | — |
case-03 | fail→fail | 26,501 | 9,798 | -63% | 1 | 1 | 0% | 3,784 | 1,376 | -64% | 0 | 0 | — |
case-04 | pass→fail | 25,799 | 43,966 | +70% | 1 | 1 | 0% | 4,017 | 1,286 | -68% | 0 | 0 | — |
case-05 | pass→fail | 18,700 | 32,466 | +74% | 1 | 1 | 0% | 2,993 | 1,335 | -55% | 0 | 0 | — |
case-06 | pass→pass | 21,351 | 26,368 | +23% | 1 | 1 | 0% | 4,290 | 4,537 | +6% | 0 | 0 | — |
case-07 | fail→fail | 10,451 | 7,672 | -27% | 1 | 1 | 0% | 1,549 | 1,268 | -18% | 0 | 0 | — |
case-09 | pass→pass | 15,301 | 5,422 | -65% | 1 | 1 | 0% | 2,487 | 1,648 | -34% | 0 | 0 | — |
case-10 | fail→pass | 16,756 | 5,265 | -69% | 1 | 1 | 0% | 2,660 | 1,683 | -37% | 0 | 0 | — |
case-11 | fail→pass | 16,257 | 6,676 | -59% | 1 | 1 | 0% | 2,514 | 1,757 | -30% | 0 | 0 | — |
case-12 | pass→pass | 15,585 | 6,901 | -56% | 1 | 1 | 0% | 2,334 | 1,815 | -22% | 0 | 0 | — |
case-13 | pass→pass | 18,076 | 14,989 | -17% | 1 | 1 | 0% | 2,250 | 2,882 | +28% | 0 | 0 | — |
case-14 | pass→pass | 17,927 | 15,176 | -15% | 1 | 1 | 0% | 2,502 | 2,998 | +20% | 0 | 0 | — |
case-15 | pass→pass | 18,339 | 16,689 | -9% | 1 | 1 | 0% | 2,664 | 3,207 | +20% | 0 | 0 | — |
case-16 | pass→pass | 18,344 | 18,207 | -1% | 1 | 1 | 0% | 2,437 | 3,281 | +35% | 0 | 0 | — |
case-17 | pass→pass | 19,641 | 24,706 | +26% | 1 | 1 | 0% | 2,533 | 4,008 | +58% | 0 | 0 | — |
case-18 | pass→pass | 21,054 | 14,505 | -31% | 1 | 1 | 0% | 2,614 | 2,853 | +9% | 0 | 0 | — |
case-19 | pass→pass | 12,832 | 10,736 | -16% | 1 | 1 | 0% | 1,894 | 2,358 | +24% | 0 | 0 | — |
case-20 | fail→fail | 8,911 | 8,302 | -7% | 1 | 1 | 0% | 1,264 | 1,297 | +3% | 0 | 0 | — |
case-21 | pass→pass | 17,357 | 12,457 | -28% | 1 | 1 | 0% | 2,323 | 2,644 | +14% | 0 | 0 | — |
case-22 | fail→fail | 27,162 | 14,363 | -47% | 1 | 1 | 0% | 4,093 | 1,539 | -62% | 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 13 counted toward the lift figure. The other 9 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 0 percentage points is the difference between those two pass rates over the 13 comparable cases. 6 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.