/godot-auditor

Aurelius Protocol: Genre Simulation NEVER List

• NEVER use floating-point for primary currency; strictly use Integer Cents (or fixed-point math) to prevent accumulated precision errors in financial models.
• NEVER process 1000+ entities individually in _process(); strictly use a Tick Manager to batch updates or process entities in rotating pools.
• NEVER rely on linear cost scaling; strictly use Exponential Growth (Base * pow(1.15, Level)) to maintain challenge and strategic tension.
• NEVER hide critical metrics from the player; strictly provide Detailed Breakdowns (Income vs. Expense) so players can make optimization-based decisions.
• NEVER allow infinite resource stacking; strictly enforce Logistical Caps (warehouses/silos) to create meaningful space-management gameplay loops.
• NEVER let the early game become a "Waiting Simulator"; strictly Front-Load Decisions and quick early wins to build player momentum.
• NEVER modify a shared Resource directly; strictly use duplicate() to avoid unintentionally updating every building of that type.
• NEVER tie simulation logic to the visual framerate; strictly use _physics_process() or delta accumulators for deterministic simulation results.
• NEVER update UI labels every frame; strictly use Event-Driven Signals to refresh UI ONLY when the underlying data changes.
• NEVER run heavy economic loops synchronously; strictly use WorkerThreadPool to offload complex calculations and prevent UI stutters.
• NEVER store massive resource data as Nodes; strictly use RefCounted or Data Resources to avoid the memory/CPU overhead of the SceneTree.
• NEVER ignore OS.low_processor_usage_mode; strictly enable it for stationary management screens to save massive CPU/Battery life.
• NEVER manipulate the SceneTree from background threads; strictly use call_deferred() for thread-safe UI updates.
• NEVER parse large JSON save files on the main thread; strictly use Threaded Serialization or optimized binary .res formats.
• NEVER use standard equality (==) for needs; strictly use is_equal_approx() to prevent floating-point jitter failures in logic gates.