This cluster-randomized study aims to evaluate the impact of a web-based clinical support tool on adherence to recommended maximum durations for benzodiazepine prescriptions among new users in primary care. General practitioners (GPs) participating in the study will have access to the website, which provides evidence-based algorithms for the management of common mental health conditions, including anxiety and sleep disorders. Prescription data and patient characteristics will be extracted from the French national health insurance database (DCIR). The primary outcome is the proportion of new benzodiazepine users whose prescriptions comply with recommended duration guidelines. Multilevel logistic regression models will assess the intervention effect, adjusting for patient- and GP-level factors.