Risk controls worth defining explicitly
Fixed size, percentage risk or another deterministic sizing model should have clear bounds and validation.
The system can restrict total exposure, per-symbol exposure or the number of simultaneous positions.
Stop Loss, Take Profit and emergency protection should have clearly defined responsibility and precedence.
Rules can prevent new trades after defined loss, trade-count or time thresholds where the strategy requires them.
A cooldown after an exit can prevent immediate re-entry when that behavior is part of the specification.
The system should define what happens when prices, account state, APIs or required data cannot be validated.
Risk logic should not depend on the entry signal
Keeping risk checks separate means every strategy decision passes through the same safety gate. This is especially useful when a system has several entry modules or trades multiple symbols.
Emergency protection is not the same as normal strategy management
A strategy may use virtual or conditional exits while a separate emergency layer exists only for abnormal conditions. Mixing both roles into one rule can make behavior harder to diagnose.
Observability is part of operational risk
If the bot rejects a trade, changes a stop, enters cooldown or triggers emergency logic, that decision should be visible in logs or alerts. A control that cannot be diagnosed is difficult to trust operationally.