Accurate assessment of children’s fundamental motor skills (FMS) is crucial for promoting lifelong healthy development and formulating effective physical education policies. However, China currently lacks standardized assessment tools that cover the entire age range from 3 to 12 years and have undergone thorough cultural adaptation. This study aimed to evaluate the psychometric properties and cultural adaptability of the MOBAK assessment tool in measuring FMS in Chinese children aged 3 to 12 years. A total of 1200 Chinese children from four regions of China participated in the study, including 623 boys (52%) and 577 girls (48%). The MOBAK tool was used to assess FMS across different age groups, focusing on two dimensions: object movement (e.g., throwing, catching, bouncing, and dribbling) and self-movement (e.g., balancing, rolling, jumping, and running). The study evaluated psychometric properties, including reliability and validity. Results indicate that MOBAK demonstrates excellent psychometric characteristics: (1) Good item discrimination (all CR values p < 0.001), with an appropriate difficulty index (0.51–0.67); (2) Extremely high reliability, manifested by high internal consistency (α > 0.80), high test–retest stability, and high inter-rater consistency (ICC > 0.90); (3) Robust construct validity, supported by exploratory and confirmatory factor analyses, which consistently confirmed the hypothesized two-factor model and had excellent fit indicators (CFI/TLI > 0.90, RMSEA/SRMR < 0.08). The MOBAK battery demonstrates strong psychometric properties and cultural validity in the Chinese context for reliably assessing FMS in children aged 3–12 years. These findings provide a foundation for future cross-cultural comparisons and validation studies in other populations.

This study presents the design, construction, and preliminary evaluation of a university-oriented digital-intelligent sports platform integrating a large language model with retrieval-augmented generation (RAG), campus data services, and domain knowledge resources. The platform architecture comprises technical, knowledge, operational-data, and user-application components supporting teaching assistance, exercise guidance, competition organization, and health-related information services. A formative campus deployment involving 428 students and six instructors documented implementation feasibility and user-facing application scenarios. To provide a more rigorous empirical supplement, an independent six-week, non-randomized, intact-class quasi-experiment was conducted with 196 students in four physical-education classes. Two classes received platform-supported instruction and two received conventional instruction from the same instructor. Baseline-adjusted analyses indicated higher posttest basic psychological need satisfaction in the platform-supported group (adjusted difference = 0.618, 95% CI 0.506–0.730, p < 0.001) and lower need frustration (adjusted difference = −0.653, 95% CI −0.759 to −0.547, p < 0.001). The adjusted difference in self-reported physical activity was not statistically significant ( p = 0.216). These findings provide preliminary evidence of more favorable psychological-need experiences during platform-supported physical education, but do not indicate a corresponding increase in self-reported physical activity over 6 weeks. The central contribution lies in the education-oriented integration of AI architecture, institutional knowledge resources, and classroom implementation evidence within a unified university physical-education platform.