Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Yahoo Finance market data downloader. Stock prices, options chains, fundamentals, dividends, splits, earnings, institutional holders, and financial statements. Quick data ingestion for quant research and backtesting.
.claude/skills/mkurman-yfinance/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -50% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-21 | ✓→✗ | ▼ Worse | -14% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -36% | 0% |
yfinance downloads Yahoo Finance market data: stock prices, options chains, fundamentals, dividends, splits, earnings, institutional holders, and financial statements. The fastest path from ticker symbol to pandas DataFrame for quant research and backtesting.
bashuv pip install yfinance
pythonimport yfinance as yf msft = yf.download("MSFT", start="2024-01-01", end="2024-12-31") print(msft.head())
pythonticker = yf.Ticker("AAPL") info = ticker.info print(f"Market cap: {info['marketCap']:,}") print(f"PE ratio: {info['trailingPE']}") print(f"Dividend yield: {info.get('dividendYield', 0)*100:.2f}%") print(ticker.balance_sheet) print(ticker.financials)
pythonopt = ticker.option_chain(ticker.options[0]) print(opt.calls[["strike", "lastPrice", "impliedVolatility", "volume"]].head()) print(opt.puts[["strike", "lastPrice", "impliedVolatility", "volume"]].head())
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,407 | 4,610 | -45% | 1 | 1 | 0% | 1,587 | 1,119 | -29% | 0 | 0 | — |
case-02 | fail→fail | 9,633 | 6,121 | -36% | 1 | 1 | 0% | 1,868 | 1,544 | -17% | 0 | 0 | — |
case-03 | pass→pass | 4,259 | 1,505 | -65% | 1 | 1 | 0% | 738 | 474 | -36% | 0 | 0 | — |
case-04 | fail→pass | 7,843 | 3,507 | -55% | 1 | 1 | 0% | 1,490 | 949 | -36% | 0 | 0 | — |
case-05 | pass→pass | 6,166 | 3,463 | -44% | 1 | 1 | 0% | 1,148 | 876 | -24% | 0 | 0 | — |
case-06 | pass→pass | 6,606 | 2,382 | -64% | 1 | 1 | 0% | 1,199 | 723 | -40% | 0 | 0 | — |
case-07 | pass→pass | 8,037 | 2,617 | -67% | 1 | 1 | 0% | 1,377 | 689 | -50% | 0 | 0 | — |
case-08 | fail→fail | 8,023 | 4,581 | -43% | 1 | 1 | 0% | 1,523 | 1,136 | -25% | 0 | 0 | — |
case-22 | fail→fail | 20,999 | 15,125 | -28% | 1 | 1 | 0% | 3,900 | 3,285 | -16% | 0 | 0 | — |
case-09 | pass→pass | 8,196 | 2,072 | -75% | 1 | 1 | 0% | 530 | 636 | +20% | 0 | 0 | — |
case-10 | fail→fail | 7,932 | 7,571 | -5% | 1 | 1 | 0% | 1,524 | 975 | -36% | 0 | 0 | — |
case-11 | pass→pass | 7,782 | 2,538 | -67% | 1 | 1 | 0% | 1,381 | 722 | -48% | 0 | 0 | — |
case-12 | pass→pass | 10,269 | 5,004 | -51% | 1 | 1 | 0% | 1,825 | 1,308 | -28% | 0 | 0 | — |
case-13 | pass→pass | 9,701 | 3,307 | -66% | 1 | 1 | 0% | 1,754 | 894 | -49% | 0 | 0 | — |
case-14 | fail→pass | 12,181 | 4,073 | -67% | 1 | 1 | 0% | 1,993 | 995 | -50% | 0 | 0 | — |
case-15 | pass→pass | 2,402 | 1,937 | -19% | 1 | 1 | 0% | 350 | 598 | +71% | 0 | 0 | — |
case-16 | pass→pass | 5,125 | 3,175 | -38% | 1 | 1 | 0% | 1,062 | 913 | -14% | 0 | 0 | — |
case-17 | pass→pass | 3,416 | 2,536 | -26% | 1 | 1 | 0% | 581 | 701 | +21% | 0 | 0 | — |
case-18 | pass→pass | 11,127 | 4,735 | -57% | 1 | 1 | 0% | 2,101 | 1,037 | -51% | 0 | 0 | — |
case-19 | pass→pass | 11,238 | 4,435 | -61% | 1 | 1 | 0% | 1,969 | 1,046 | -47% | 0 | 0 | — |
case-20 | fail→pass | 23,763 | 11,238 | -53% | 1 | 1 | 0% | 4,537 | 2,324 | -49% | 0 | 0 | — |
case-21 | pass→fail | 22,436 | 17,065 | -24% | 1 | 1 | 0% | 4,006 | 3,445 | -14% | 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. The headline lift of +9 percentage points is the difference between those two pass rates over the 22 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.