Structural pathologies, such as cracks, spalling concrete, and corrosion, pose significant challenges in civil engineering (CE), threatening the safety and longevity of infrastructures. Traditional inspection methods rely heavily on manual expertise, are time-intensive, and often fail to detect defects at an early stage. These limitations underscore the need for innovative automated solutions. To address these limitations, we developed Aigis, an AI-based system designed to detect and diagnose structural pathologies. Leveraging a Convolutional Neural Network (CNN) trained on annotated images, Aigis identifies critical defects with high accuracy. To ensure its practical applicability, the system integrates international engineering standards (such Eurocodes) to classify the severity of detected pathologies. This combination of AI and engineering norms transforms Aigis from a mere detection tool into a comprehensive diagnostic system. The methodology involves data collection and annotation using Label Studio, model training with a balanced dataset of real pathology images, and integration of Eurocode thresholds to enhance diagnostic accuracy. Initial tests yielded a validation accuracy of 87%, with promising results observed in real-world evaluations. Aigis demonstrates significant potential for scalability and efficiency. By automating pathology diagnostics, it reduces reliance on manual inspections, enhances structural safety, and promotes preventive maintenance.