Cultural heritage masonry structures are particularly vulnerable to various damage mechanisms, such as ground deformation-induced settlements, seismic cracking, and long-term material degradation. Their unique architectural and material characteristics make their assessment and preservation a complex task, especially within the dynamic context of urban environments. In such settings, the failure of one building can have cascading effects on neighboring structures, while shared threats often impact multiple buildings simultaneously. To address these challenges, this study proposes a Bayesian Network (BN)-based methodology to assist decision makers for a risk-informed intervention plan at the urban scale. The framework integrates diverse data sources to be analyzed and compared, including historical documentation, and expert knowledge, to quantify failure probabilities across multiple scenarios and hazard conditions. By incorporating dependencies among structural vulnerability, failure mechanisms, and available monitoring systems, the BN approach ensures a robust and adaptable process to assist decision making. To demonstrate the efficacy of the proposed methodology, a case study is selected as the source domain, and a calibrated digital twin is used for model updating and physics-based damage assessment. The information acquired from the source domain is then used to infer the risk-based knowledge in other target domains at the urban scale.  The proposed methodology has the potential to evaluate risks associated with differential settlements, excavation activities, earthquakes, and progressive deterioration. The main goal is to demonstrate how the scalable BN-based approach can provide actionable insights for urban planners and decision-makers, enabling targeted interventions and risk-informed urban management.