Make Risk Permission Impossible to Bypass
Earlier modules translated the source's 'trade from risk' principle into technical stop validation, risk budget and position sizing. In production architecture, those rules need a hard software boundary.
Architecture principle
The risk layer receives TRADE_INTENT plus account and portfolio state. It returns SAFE, REDUCE or BLOCK together with approved volume, stop parameters and reason codes. These permission states are an FX Nova Bot engineering formalization rather than terminology from the source.
The execution adapter accepts only an APPROVED_ORDER object produced by risk. A raw strategy signal should be structurally incapable of reaching the broker API.
Connected FX Nova modules
Educational software-engineering material. Architecture beyond the source trading methodology is explicitly presented as FX Nova Bot engineering design.