Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate trading signals using technical indicators (RSI, MACD, Bollinger Bands, etc.). Combines multiple indicators into composite signals with confidence scores. Use when analyzing assets for trading opportunities or checking technical indicators. Trigger with phrases like "get trading signals", "check indicators", "analyze for entry", "scan for opportunities", "generate buy/sell signals", or "technical analysis".
.claude/skills/jeremylongshore-generating-trading-signals/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -50% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 1% | 0% |
Multi-indicator signal generation system that analyzes price action using 7 technical indicators and produces composite BUY/SELL signals with confidence scores and risk management levels.
Indicators: RSI, MACD, Bollinger Bands, Trend (SMA 20/50/200), Volume, Stochastic Oscillator, ADX.
Install required dependencies:
bashset -euo pipefail pip install yfinance pandas numpy
Optional for visualization: pip install matplotlib
bash python ${CLAUDE_SKILL_DIR}/scripts/scanner.py --watchlist crypto_top10 --period 6m
Output shows signal type (STRONG_BUY/BUY/NEUTRAL/SELL/STRONG_SELL) and confidence per asset.
bash python ${CLAUDE_SKILL_DIR}/scripts/scanner.py --symbols BTC-USD --detail
Shows each indicator's individual signal, value, and reasoning.
bash # Only buy signals with 70%+ confidence python ${CLAUDE_SKILL_DIR}/scripts/scanner.py --filter buy --min-confidence 70 --rank confidence
# Save results to JSON python ${CLAUDE_SKILL_DIR}/scripts/scanner.py --output signals.json
bash python ${CLAUDE_SKILL_DIR}/scripts/scanner.py --list-watchlists python ${CLAUDE_SKILL_DIR}/scripts/scanner.py --watchlist crypto_defi
Available: crypto_top10, crypto_defi, crypto_layer2, stocks_tech, etfs_major
The scanner produces a summary table with symbol, signal type, confidence %, price, and stop loss for each asset scanned. Detailed mode adds per-indicator breakdowns with risk management levels (stop loss, take profit, risk/reward ratio).
Signal types: STRONG_BUY (+2), BUY (+1), NEUTRAL (0), SELL (-1), STRONG_SELL (-2)
Confidence ranges: 70-100% high conviction | 50-70% moderate | 30-50% weak | 0-30% avoid
See ${CLAUDE_SKILL_DIR}/references/implementation.md for full output format examples and signal type tables.
| Error | Cause | Fix | |-------|-------|-----| | No data for symbol | Invalid ticker or delisted | Verify symbol exists on Yahoo Finance | | Insufficient data | Period too short for indicators | Use --period 6m minimum | | Rate limit exceeded | Too many rapid API calls | Add delay between scans |
See ${CLAUDE_SKILL_DIR}/references/errors.md for comprehensive error handling.
Morning crypto scan - Check all top-10 crypto assets for entry opportunities:
bashpython ${CLAUDE_SKILL_DIR}/scripts/scanner.py --watchlist crypto_top10 --period 6m
Deep dive on Bitcoin - Full indicator breakdown with risk management levels:
bashpython ${CLAUDE_SKILL_DIR}/scripts/scanner.py --symbols BTC-USD --detail
Find strongest DeFi buy signals - Filter and rank by confidence:
bashpython ${CLAUDE_SKILL_DIR}/scripts/scanner.py --watchlist crypto_defi --filter buy --rank confidence
Export results - Save to JSON for automated pipeline or further analysis:
bashpython ${CLAUDE_SKILL_DIR}/scripts/scanner.py --watchlist crypto_top10 --output signals.json
${CLAUDE_SKILL_DIR}/references/implementation.md - Output formats, configuration, backtester integration, file reference| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 19,395 | 9,999 | -48% | 1 | 1 | 0% | 2,588 | 1,265 | -51% | 0 | 0 | — |
case-02 | fail→fail | 29,222 | 18,527 | -37% | 1 | 1 | 0% | 5,236 | 1,289 | -75% | 0 | 0 | — |
case-03 | fail→fail | 29,371 | 12,709 | -57% | 1 | 1 | 0% | 5,472 | 1,227 | -78% | 0 | 0 | — |
case-04 | fail→pass | 14,983 | 7,549 | -50% | 1 | 1 | 0% | 2,769 | 1,390 | -50% | 0 | 0 | — |
case-05 | fail→fail | 14,224 | 7,644 | -46% | 1 | 1 | 0% | 1,422 | 1,452 | +2% | 0 | 0 | — |
case-06 | fail→pass | 13,783 | 7,601 | -45% | 1 | 1 | 0% | 1,461 | 1,449 | -1% | 0 | 0 | — |
case-07 | fail→pass | 17,238 | 8,190 | -52% | 1 | 1 | 0% | 2,195 | 1,494 | -32% | 0 | 0 | — |
case-08 | pass→pass | 7,082 | 7,065 | -0% | 1 | 1 | 0% | 1,257 | 1,315 | +5% | 0 | 0 | — |
case-09 | fail→pass | 6,392 | 7,020 | +10% | 1 | 1 | 0% | 1,096 | 1,291 | +18% | 0 | 0 | — |
case-10 | pass→pass | 12,965 | 2,209 | -83% | 1 | 1 | 0% | 1,413 | 1,331 | -6% | 0 | 0 | — |
case-11 | pass→pass | 8,793 | 6,602 | -25% | 1 | 1 | 0% | 1,744 | 1,206 | -31% | 0 | 0 | — |
case-12 | pass→pass | 6,239 | 6,397 | +3% | 1 | 1 | 0% | 1,135 | 1,210 | +7% | 0 | 0 | — |
case-13 | pass→pass | 5,219 | 1,394 | -73% | 1 | 1 | 0% | 926 | 1,169 | +26% | 0 | 0 | — |
case-14 | fail→pass | 7,661 | 2,836 | -63% | 1 | 1 | 0% | 1,389 | 1,396 | +1% | 0 | 0 | — |
case-15 | pass→pass | 14,234 | 2,123 | -85% | 1 | 1 | 0% | 1,783 | 1,323 | -26% | 0 | 0 | — |
case-16 | pass→pass | 12,744 | 6,962 | -45% | 1 | 1 | 0% | 1,571 | 1,312 | -16% | 0 | 0 | — |
case-17 | pass→pass | 7,676 | 6,498 | -15% | 1 | 1 | 0% | 1,653 | 1,239 | -25% | 0 | 0 | — |
case-18 | pass→pass | 10,599 | 9,012 | -15% | 1 | 1 | 0% | 1,982 | 1,817 | -8% | 0 | 0 | — |
case-19 | fail→pass | 6,921 | 1,631 | -76% | 1 | 1 | 0% | 334 | 1,289 | +286% | 0 | 0 | — |
case-20 | fail→pass | 13,891 | 6,553 | -53% | 1 | 1 | 0% | 1,816 | 1,267 | -30% | 0 | 0 | — |
case-21 | fail→fail | 15,419 | 13,779 | -11% | 1 | 1 | 0% | 3,586 | 4,049 | +13% | 0 | 0 | — |
case-22 | fail→fail | 12,799 | 12,886 | +1% | 1 | 1 | 0% | 2,223 | 3,648 | +64% | 0 | 0 | — |
case-23 | pass→pass | 23,431 | 21,911 | -6% | 1 | 1 | 0% | 3,418 | 4,094 | +20% | 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. 23 cases were attempted, and 20 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 +30 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.