Scan US large-cap stocks for Connors RSI(2) oversold signals in uptrends.
npx skills add https://github.com/mthli/skills --skill mean-reversion-scan
Please help me install this Agent Skill. Skill: mean-reversion-scan Source: https://github.com/mthli/skills/tree/main/mean-reversion-scan Command: npx skills add https://github.com/mthli/skills --skill mean-reversion-scan
It helps you identify mean-reversion equity opportunities by scanning for Connors-style RSI(2) oversold conditions that occur inside confirmed long-term uptrends, so you can focus on potential “buy-the-dip” setups rather than manually checking charts.
Run mean-reversion-scan with the standard RSI(2) trigger to list the top currently triggered oversold bounce candidates for liquid US large-cap stocks.
High-intent search queries and answers about installing and using this skill.
To scan for RSI(2) oversold bounces, the skill screens US large-cap equities using Connors RSI(2) signals gated by 200DMA and 50DMA trend filters. It identifies potential buy-the-dip setups where the long-term uptrend remains intact, ensuring you only find bounce candidates in healthy trends.
Mean-reversion screening with market regime filters detects short-term oversold reversals while confirming a broader risk-on environment using SPY vs its 200DMA. This dual-layer approach prevents buying dips during systemic risk-off periods, focusing only on high-probability bounce entries.
Yes, you can use yfinance OHLCV data for multi-ticker equity screening within this skill. It retrieves historical price data via yfinance to compute RSI(2), calculate ATR-based stops, define 5DMA targets, and resolve historical win rates for liquid US large-cap stocks.
ATR stops and 5DMA targets are calculated directly from yfinance OHLCV data to provide practical execution levels for each mean-reversion candidate. The ATR-based stop defines downside risk, while the 5DMA serves as the profit target for resolving win-rate statistics.
Yes, mean-reversion backtesting tracks persistent oversold non-bouncing stocks by logging prior signals in a history file. It features a dedicated section for names stuck in oversold territory, separating them from successful bounces to maintain accurate running win-rate statistics.
A limitation of using RSI(2) for large-cap mean-reversion is the risk of remaining stuck in oversold territory during trend breakdowns. While 200DMA and market regime filters mitigate this, the scanner explicitly isolates persistent non-bouncing names to highlight unresolved risk.
koukoCompute technical indicators from OHLCV data into structured snapshot cards.
gracefullightBacktest trading strategies against historical data and export performance metrics.
jayjpatel9717Automate portfolio allocation and risk management with Python calculations.
derekcrossluInterpret backtest results and detect overfitting in trading strategies.
