Abstract: In the context of urban stock development, the matching of supply and demand as well as the equitable allocation of green spaces with disaster prevention and emergency evacuation functions have become key issues affecting urban safety. Taking the central urban area of Chengdu as an example and using streets as the basic units, this study constructs an index system covering four dimensions of green space supply (quantity allocation, transportation organization, emergency services, and spatial layout) and three dimensions of social demand (population density, disaster risk, and built environment). By applying a supply–demand coupling coordination model, the spatial differentiation characteristics and fairness pattern of disaster prevention green spaces are systematically evaluated. The results indicate that: (1) Both the supply and demand of disaster prevention green spaces in central Chengdu show a "high in the center, low in the periphery" ring-shaped distribution pattern. (2) The supply–demand matching types are dominated by high-level balance in central areas and low-level balance in outer areas. The overall coupling coordination degree is relatively poor, with most streets in a state of primary or barely coordinated development. (3) Spatial fairness varies significantly across regions, with certain northern streets experiencing prominent fairness deficits due to lagging supply. The findings provide a quantitative basis for optimizing urban safety patterns and precisely allocating disaster prevention resources, thereby contributing to the balanced distribution of green space benefits and the high-quality, coordinated development of urban systems.