The balance between ecosystem service (ES) supply and demand underpins ecological security, making the construction of ecological security patterns (ESPs) integrating ES supply-demand dynamics with human well-being critical for regional sustainability. However, existing studies often focus on single administrative units or static assessments while oversimplifying ecological resistance surfaces, inadequately reflecting the complexity of securing ecological well-being and resulting in ESPs with limited resilience and adaptability. To address this, we propose a novel method for constructing social-ecological security patterns by coupling ES flows with supply-demand ratios. Applying this approach to China’s ecologically fragile Loess hilly and gully region, we integrated the Gaussian two-step floating catchment area method and circuit theory to refine traditional resistance surfaces, establish dedicated supply-demand corridors and nodes, and formulate a holistic, synergistic security pattern. The results revealed: (1) a spatially heterogeneous ES supply ("high southeast, low northwest") contrasting with demand exhibiting "linear clustering and scattered distribution", where water conservation and soil retention faced the most severe deficits; (2) identification of 67 ecological source areas (10,776.87 km2), 33 demand source areas (1,632.27 km2), 7,480.64 km of ecological corridors (demonstrating higher network efficiency in the southeast), 8,991.5 km of supply-demand corridors, 82 ecological nodes, and 9 supply-demand nodes; and (3) development of a spatial framework termed "four belts, four zones, one barrier, and three clusters", accompanied by targeted conservation-restoration strategies to reconcile ecological protection with economic development. This study provides a scientific foundation for advancing sustainable economic growth and human well-being in the Loess hilly and gully region.