French researchers implemented an automated whole-genome sequencing pipeline that identified 75 tuberculosis transmission clusters among 1,152 patients across eight hospitals from 2016-2025, affecting 21.4% of cases. The system provided real-time alerts to public health teams and accurately predicted drug resistance using WHO mutation catalogues. Unexpectedly, traditional contagiousness markers like positive sputum smears and lung cavities showed no association with extra-household transmission rates. Instead, patients with lower disease severity and longer symptom duration before diagnosis drove more community spread. This counterintuitive finding challenges conventional TB control assumptions and suggests that less severely ill patients may unknowingly spread infection longer before seeking care. The automated approach enabled clinical microbiologists to interpret genomic data directly, facilitating immediate public health responses including investigation of school outbreaks and hospital transmissions. While promising for TB surveillance in low-burden countries, this preprint awaits peer review and the findings require validation in diverse healthcare settings. The work represents an incremental but practically significant advance in applying genomics to infectious disease control, potentially reshaping how health systems prioritize contact tracing resources.