simada: A Behavioral Simulation Framework for Type 1 Diabetes Automated Insulin Delivery Research (v0.1.0)

 simada (Simulation of AID Adherence) is an open-source Python framework for Type 1 Diabetes behavioral simulation, built on top of simglucose (an open-source Python implementation of the UVA/Padova T1D metabolic model). simada extends simglucose with behavioral archetypes (adherent/moderate/nonadherent), culturally-authentic Brazilian dietary patterns via the TACO food composition table, pharmacological adjunct modeling (semaglutide PK/PD), and a clinical-metrics analysis pipeline following the International Consensus on Time in Range (Battelino et al., 2019). This release (v0.1.0) is the initial snapshot of the framework prior to upcoming publications. The framework is intended for Safe Reinforcement Learning research on Automated Insulin Delivery (AID) systems and for in silico exploration of behavioral and  pharmacological factors in T1D control.