Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Search A-Share/HK/US finance stock tickers and retrieve finance stock price history. Use when user asks about finance stock codes, recent price changes, or specific company finance stock info.
.claude/skills/rkiding-alphaear-stock/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -58% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -47% | 0% |
Search A-Share/HK/US stock tickers and retrieve historical price data (OHLCV).
Use scripts/stock_tools.py via StockTools.
search_ticker(query){code, name}.get_stock_price(ticker, start_date, end_date)get_stock_fundamentals(ticker)pandas, requests, akshare, yfinancescripts/database_manager.py (stock tables)yfinance), you may need to set environment variables if your network cannot reach Yahoo Finance directly:bash export HTTP_PROXY="http://<proxy_ip>:<port>" export HTTPS_PROXY="http://<proxy_ip>:<port>"
akshare (EastMoney), which usually works best with a direct connection in China. The tool automatically detects proxy issues and attempts direct connection for these markets.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 9,140 | 5,964 | -35% | 1 | 1 | 0% | 2,154 | 758 | -65% | 0 | 0 | — |
case-02 | fail→fail | 24,209 | 7,391 | -69% | 1 | 1 | 0% | 5,232 | 624 | -88% | 0 | 0 | — |
case-03 | fail→pass | 4,872 | 2,474 | -49% | 1 | 1 | 0% | 827 | 814 | -2% | 0 | 0 | — |
case-04 | fail→pass | 11,601 | 3,119 | -73% | 1 | 1 | 0% | 2,730 | 1,106 | -59% | 0 | 0 | — |
case-05 | pass→pass | 10,964 | 2,776 | -75% | 1 | 1 | 0% | 2,165 | 864 | -60% | 0 | 0 | — |
case-06 | fail→pass | 7,076 | 2,116 | -70% | 1 | 1 | 0% | 1,349 | 751 | -44% | 0 | 0 | — |
case-07 | pass→pass | 5,460 | 1,195 | -78% | 1 | 1 | 0% | 1,000 | 564 | -44% | 0 | 0 | — |
case-17 | pass→pass | 12,408 | 2,618 | -79% | 1 | 1 | 0% | 2,566 | 813 | -68% | 0 | 0 | — |
case-08 | fail→pass | 13,100 | 3,964 | -70% | 1 | 1 | 0% | 2,586 | 1,094 | -58% | 0 | 0 | — |
case-09 | fail→pass | 6,671 | 1,694 | -75% | 1 | 1 | 0% | 1,214 | 638 | -47% | 0 | 0 | — |
case-10 | fail→pass | 11,075 | 2,897 | -74% | 1 | 1 | 0% | 2,354 | 868 | -63% | 0 | 0 | — |
case-11 | pass→pass | 5,119 | 1,271 | -75% | 1 | 1 | 0% | 862 | 539 | -37% | 0 | 0 | — |
case-12 | pass→pass | 6,216 | 1,379 | -78% | 1 | 1 | 0% | 1,081 | 554 | -49% | 0 | 0 | — |
case-13 | fail→pass | 8,630 | 2,533 | -71% | 1 | 1 | 0% | 1,828 | 810 | -56% | 0 | 0 | — |
case-14 | pass→pass | 7,138 | 1,541 | -78% | 1 | 1 | 0% | 1,274 | 601 | -53% | 0 | 0 | — |
case-15 | fail→pass | 5,279 | 2,246 | -57% | 1 | 1 | 0% | 951 | 748 | -21% | 0 | 0 | — |
case-16 | pass→pass | 8,419 | 2,052 | -76% | 1 | 1 | 0% | 1,577 | 760 | -52% | 0 | 0 | — |
case-18 | pass→pass | 9,939 | 3,173 | -68% | 1 | 1 | 0% | 1,801 | 989 | -45% | 0 | 0 | — |
case-19 | pass→pass | 9,674 | 2,520 | -74% | 1 | 1 | 0% | 1,976 | 852 | -57% | 0 | 0 | — |
case-20 | pass→pass | 2,596 | 3,279 | +26% | 1 | 1 | 0% | 434 | 856 | +97% | 0 | 0 | — |
case-21 | fail→fail | 18,293 | 7,956 | -57% | 1 | 1 | 0% | 3,738 | 880 | -76% | 0 | 0 | — |
case-22 | fail→fail | 7,287 | 19,655 | +170% | 1 | 1 | 0% | 1,508 | 3,515 | +133% | 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 19 counted toward the lift figure. The other 3 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 +36 percentage points is the difference between those two pass rates over the 19 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.