<jats:p>Restoration aims to return ecosystem services, including the human health benefits of exposure to green space. The loss of such exposure with urbanization and industrialization has arguably contributed to an increase in human immune dysregulation. The Biodiversity and Old Friends hypotheses have described the possible mechanisms of this relationship, and suggest that reduced exposure to diverse, beneficial microorganisms can result in negative health consequences. However, it is unclear whether restoration of biodiverse habitat can reverse this effect, and what role the environmental microbiome might have in such recovery. Here, we propose the <jats:italic>Microbiome Rewilding Hypothesis</jats:italic>, which specifically outlines that restoring biodiverse habitats in urban green spaces can rewild the environmental microbiome to a state that enhances primary prevention of human disease. We support our hypothesis with examples from allied fields, including a case study of active restoration that reversed the degradation of the soil bacterial microbiome of a former pasture. This case study used high‐throughput amplicon sequencing of environmental DNA to assess the quality of a restoration intervention in restoring the soil bacterial microbiome. The method is rapid, scalable, and standardizable, and has great potential as a monitoring tool to assess functional outcomes of green‐space restoration. Evidence for the Microbiome Rewilding Hypothesis will help motivate health professionals, urban planners, and restoration practitioners to collaborate and achieve co‐benefits. Co‐benefits include improved human health outcomes and investment opportunities for biodiversity conservation and restoration.</jats:p>