Early identification of scoliosis is highly significant for timely intervention and efficient management of the disease. However, conventional static screening approaches rely heavily on subjective assessments, resulting in limited diagnostic accuracy and insufficient continuity. Here, we propose a motion vector sensor-enabled 3D spinal morphology monitoring suit for intelligent identification and continuous monitoring of scoliosis. By orderly aligning elastic conductive fibers, highly anisotropic soft meshes are customized and achieve dual responses to motion amplitude and direction. The sensors exhibit a minimal directional resolution of 2° while maintaining a linear response up to 100% strain. The suit integrates motion sensors positioned at symmetrical body locations to quantify 3D spinal motions via precise measurement of skin-surface deformation amplitude and direction. Dynamic scoliosis identification is achieved through machine learning-powered kinematic symmetry evaluation of 3D spinal motions. The system demonstrates accuracies of 100% for scoliosis detection, 97.7% for curvature degree quantification, and 93.37% for curvature location identification, rigorously validated on a multi-center dataset from four hospitals and one school-based screening program. This work provides a radiation-free alternative for scoliosis screening, with potential to facilitate lifelong monitoring and management of the condition.