Dysbiosis is a hallmark of atopic dermatitis (AD). The composition of skin microbiome communities and the causality of dysbiosis in eczema have not been well established. The objective of this review is to describe the skin microbiome profile in AD and address whether there is a causal relationship between dysbiosis and AD. The protocol is registered in PROSPERO (CRD42016035813). We searched PubMed, Embase, Scopus and ClinicalTrials.gov for primary research studies applying culture-independent analysis on the microbiome on AD skin of humans and animal models. Two authors independently screened the full text of studies for eligibility and assessed risk of bias. Because of heterogeneity no quantitative synthesis was done. Of 5735 texts, 32 met the inclusion criteria (17 published: 11 human and six animal studies). The studies varied in quality and applied different methodology. The skin in AD had low bacterial diversity (lowest at dermatitis-involved sites) and three studies showed depletion of Malassezia spp. and high non-Malassezia fungal diversity. The relative abundance of Staphylococcus aureus and Staphylococcus epidermidis were elevated and other genera were reduced, including Propionibacterium. A mouse study indicated that dysbiosis is a driving factor in eczema pathogenesis. The data are not sufficiently robust for good characterization; however, dysbiosis in AD not only implicates Staphylococcus spp., but also microbes such as Propionibacterium and Malassezia. A causal role of dysbiosis in eczema in mice should encourage future studies to investigate if this also applies to humans. Other important aspects are temporal dynamics and the influence of methodology on microbiome data.