Run multi-indicator technical analysis on crypto or market OHLCV data. Use for deterministic trend, momentum, volume, and divergence analysis.
Detected risks:
crypto-ta-analyzer runs deterministic, multi-indicator technical analysis on cryptocurrency or general market OHLCV data. It targets the problem of narrative, one-number market opinions by producing an explicit indicator stack with signal scoring, so users get reproducible trend, momentum, volume, volatility, and divergence readouts rather than a single unexplained verdict.
The skill bundles Python scripts built on pure numpy and pandas (no TA-Lib dependency): data_converter.py and coingecko_converter.py normalize OHLCV data from sources like CoinGecko, exchanges, and Yahoo, and ta_analyzer.py computes 29 indicators (including RSI, MACD, Bollinger Bands, OBV, Ichimoku with crypto-optimized periods, VWAP, and ATR) and emits 7-tier trading signals from STRONG_BUY to STRONG_SELL. Outputs include a confidence score, RSI/MACD/OBV divergence detection, Bollinger Band squeeze alerts, and volume-confirmation scoring. The skill's guardrails explicitly instruct that signals must not be presented as guaranteed outcomes and that deterministic indicator output should be kept distinct from discretionary interpretation.
It is intended for traders, quantitative hobbyists, and analysts who want transparent, explainable technical readouts they can reason about. Note that one documentation snippet demonstrates fetching public CoinGecko price data with TLS certificate verification disabled; this is a test convenience rather than a core capability, but it is worth flagging as insecure network practice.
Normalized OHLCV (open/high/low/close/volume) data. Converter scripts reshape common source formats from CoinGecko, exchanges, or Yahoo into the expected shape before analysis.
Only numpy (>=1.21.0) and pandas (>=1.3.0). All 29 indicator algorithms are implemented in pure numpy/pandas with no TA-Lib dependency.
A 7-tier trade signal (STRONG_BUY to STRONG_SELL), a 0-1 confidence score, a boolean high-conviction trade trigger, RSI/MACD/OBV divergence detection, a Bollinger Band squeeze flag, and a volume-confirmation score.
No. The skill's guardrails require signals not to be presented as guaranteed outcomes and stress keeping deterministic indicator output separate from discretionary interpretation.
One example snippet fetches public CoinGecko data with TLS certificate verification disabled. Prefer keeping certificate validation on when adapting the fetch example for real use.
Quick Setup:
.claude/skills/Repository
dkyazzentwatwa/chatgpt-skills