Understanding aboveground-belowground linkages and their consequences for ecosystem functioning is a major challenge in soil ecology. It is already well established that soil communities drive essential ecosystem processes, such as nutrient cycling, decomposition, or carbon storage. However, knowledge of how plant diversity affects belowground community structure is limited. Such knowledge can be gained from studying the main plant functional traits that modulate plant community effects on soil fauna. Here, we used a grassland experiment manipulating plant species richness and plant functional diversity to explore the effects of community-level plant traits on soil meso- and macrofauna and the trophic structure of soil fauna by differentiating predators and prey. The functional composition of plant communities was described by six plant traits related to spatial and temporal resource use: plant height, leaf area, rooting depth, root length density, growth start, and flowering start. Community-Weighted Means (CWMs), Functional Dissimilarity (FDis), and Functional Richness (FRic) were calculated for each trait. Community-level plant traits better explained variability in soil fauna than did plant species richness. Notably, each soil fauna group was affected by a unique set of plant traits. Moreover, the identity of plant traits (CWM) explained more variance of soil fauna groups than trait diversity. The abundances of soil fauna at the lower trophic levels were better explained by community-level plant traits than higher trophic levels soil fauna groups. Taken together, our results highlight the importance of the identity of different plant functional traits in driving the diversity and trophic structure of soil food communities.