Install any skill in seconds. Free to start, no credit card required.
Get Started Free →股票个股分析,实时获取价格涨跌幅,计算技术指标和支撑位,识别缺口并判断支撑压力,智能预测未来3天走势并给出操作建议
.claude/skills/anbeime-stock-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-12 | ✓→✓ | = Same ✓ | -5% | 0% |
| case-22 | ✓→✓ | = Same ✓ | 186% | 0% |
requests>=2.28.0 numpy>=1.24.0 pandas>=2.0.0
scripts/fetch_stock_data.py 获取实时行情和历史K线数据--stock_code: 股票代码--days: 获取历史数据天数(默认30天)scripts/analyze_stock.py 进行技术分析--data_file: 上一步获取的数据文件路径用户:分析000001平安银行
执行:
1. 调用 fetch_stock_data.py --stock_code 000001 --days 30
2. 调用 analyze_stock.py --data_file stock_data_000001.json
3. 基于分析结果生成走势预测和操作建议用户:分析腾讯控股 00700.HK
执行:
1. 调用 fetch_stock_data.py --stock_code 00700.HK --days 30
2. 调用 analyze_stock.py --data_file stock_data_00700.HK.json
3. 生成分析报告和操作建议用户:分析AAPL苹果公司
执行:
1. 调用 fetch_stock_data.py --stock_code AAPL --days 30
2. 调用 analyze_stock.py --data_file stock_data_AAPL.json
3. 提供全面的技术分析报告| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,319 | 22,624 | +11% | 1 | 1 | 0% | 3,568 | 1,623 | -55% | 0 | 0 | — |
case-02 | fail→fail | 22,533 | 5,422 | -76% | 1 | 1 | 0% | 3,851 | 1,606 | -58% | 0 | 0 | — |
case-12 | pass→pass | 10,800 | 2,657 | -75% | 1 | 1 | 0% | 1,972 | 1,867 | -5% | 0 | 0 | — |
case-22 | pass→pass | 4,701 | 6,760 | +44% | 1 | 1 | 0% | 850 | 2,429 | +186% | 0 | 0 | — |
case-03 | fail→pass | 14,757 | 5,338 | -64% | 1 | 1 | 0% | 3,010 | 2,459 | -18% | 0 | 0 | — |
case-04 | fail→pass | 18,326 | 7,756 | -58% | 1 | 1 | 0% | 3,813 | 2,876 | -25% | 0 | 0 | — |
case-05 | pass→pass | 10,513 | 3,295 | -69% | 1 | 1 | 0% | 1,991 | 1,950 | -2% | 0 | 0 | — |
case-06 | pass→pass | 11,403 | 9,670 | -15% | 1 | 1 | 0% | 1,643 | 2,784 | +69% | 0 | 0 | — |
case-07 | pass→pass | 8,445 | 10,960 | +30% | 1 | 1 | 0% | 1,373 | 2,992 | +118% | 0 | 0 | — |
case-08 | fail→fail | 13,312 | 8,548 | -36% | 1 | 1 | 0% | 2,154 | 2,788 | +29% | 0 | 0 | — |
case-09 | pass→pass | 16,842 | 8,729 | -48% | 1 | 1 | 0% | 2,689 | 2,831 | +5% | 0 | 0 | — |
case-10 | pass→pass | 11,687 | 5,123 | -56% | 1 | 1 | 0% | 1,869 | 2,253 | +21% | 0 | 0 | — |
case-11 | fail→pass | 7,574 | 1,348 | -82% | 1 | 1 | 0% | 1,291 | 1,499 | +16% | 0 | 0 | — |
case-13 | pass→pass | 7,923 | 4,603 | -42% | 1 | 1 | 0% | 1,567 | 2,218 | +42% | 0 | 0 | — |
case-14 | pass→pass | 8,239 | 3,069 | -63% | 1 | 1 | 0% | 1,525 | 1,877 | +23% | 0 | 0 | — |
case-15 | pass→pass | 7,943 | 2,835 | -64% | 1 | 1 | 0% | 1,388 | 1,842 | +33% | 0 | 0 | — |
case-16 | pass→pass | 16,806 | 13,723 | -18% | 1 | 1 | 0% | 2,576 | 3,371 | +31% | 0 | 0 | — |
case-17 | pass→pass | 16,523 | 15,576 | -6% | 1 | 1 | 0% | 2,546 | 3,695 | +45% | 0 | 0 | — |
case-18 | pass→pass | 5,439 | 3,266 | -40% | 1 | 1 | 0% | 831 | 1,891 | +128% | 0 | 0 | — |
case-19 | fail→fail | 25,647 | 26,804 | +5% | 1 | 1 | 0% | 5,967 | 7,509 | +26% | 0 | 0 | — |
case-20 | fail→fail | 17,339 | 18,990 | +10% | 1 | 1 | 0% | 2,939 | 4,411 | +50% | 0 | 0 | — |
case-21 | fail→fail | 16,279 | 22,546 | +38% | 1 | 1 | 0% | 3,368 | 6,147 | +83% | 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 20 counted toward the lift figure. The other 2 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 20 comparable cases.
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.