Predatory myxobacteria are important soil micropredators with the potential to regulate soil microbial community structure and ecosystem function. However, the biogeographic distribution patterns, assembly processes, and potential nutrient cycling functions of myxobacteria communities in typical agricultural soils in China are still poorly understood. High-throughput sequencing, phylogenetic zero modeling, and the multi-nutrient cycling index were used to assess the biogeographic distribution, assembly processes, and soil ecosystem functions of predation myxobacteria communities in typical agricultural soils of six long-term fertilization ecological experimental stations. The results demonstrated a hump-shaped distribution of myxobacteria α-diversity along the latitudinal gradient and significant differences in myxobacteria β-diversity in typical agricultural soils (P < 0.05). Bacterial richness, soil organic carbon, and pH were the most important predictors of myxobacteria α-diversity, whereas geographic factors and soil pH were the most significant ecological predictors of myxobacteria β-diversity. Myxobacteria community assembly is dominated by deterministic processes, especially homogeneous selection, primarily driven by soil pH and bacterial richness. In addition, we revealed the ecological significance of myxobacteria communities in typical agricultural soil microbial networks and the potential link between myxobacteria communities and soil nutrient cycling. These findings enhance our understanding of the biogeographic distribution, community assembly, ecological predictors, and relationships with soil nutrient cycling of myxobacteria communities in typical agricultural soils, paving the way for a more predictive understanding of the effect of predatory myxobacteria communities on soil ecosystem function, which is essential for the development of sustainable agriculture.