Tune regime weights and multi-timeframe logic in QuantEngine composite signals.
npx skills add https://github.com/bitandbytes/Argus --skill quant-engine-dev
Please help me install this Agent Skill. Skill: quant-engine-dev Source: https://github.com/bitandbytes/Argus/tree/main/.claude/skills/quant-engine-dev Command: npx skills add https://github.com/bitandbytes/Argus --skill quant-engine-dev
The Quant Engine Development Skill guides engineers to modify the QuantEngine — the heart of the trading pipeline — to tune composite signals, regime-specific weights, and multi-timeframe confirmation, enabling more reliable and explainable trading decisions.
To start, load your backtest data and run a minimal QuantEngine workflow with a sample config to observe how changes affect the signal and confidence.
High-intent search queries and answers about installing and using this skill.
Tune regime weights for composite trading signals by modifying the QuantEngine signal generation flow to adjust how individual indicators contribute per market regime, ensuring the composite remains within [-1, +1] and confidence stays within [0, 1].
Multi-timeframe confirmation in a Python trading system works by adjusting regime-dependent weights across different timeframes to validate composite signals, improving accuracy and stability without using future data.
Debug unexpected signal outputs across different market regimes by running unit tests and backtests with a minimal QuantEngine workflow and sample config to observe how regime-aware parameter changes affect the final signal.
You can add support for new indicators without modifying downstream trading pipeline layers by extending the QuantEngine signal flow and plugin registry, isolating upstream changes from downstream execution.
Customize per stock overrides for regime-aware parameterization by adjusting the QuantEngine configuration to balance precision and recall for individual assets while maintaining the composite signal bounds.
Limitations of modifying composite signal calculations include strict constraints where the composite must remain within [-1, +1], confidence must stay within [0, 1], and no future data can be used in the signal generation flow.
BankrBotInitialize BankrClient with payment and context wallets from environment variables.
IgorGanapolskyAggregate news, social, and market sentiment into composite scores for trading decisions.
nrwlDiscover and install Nx plugins for monorepo frameworks.
GhostOf0daysAutomate capacity crowding workflows for quantitative trading systems.
