Inhibition of human epidermal growth factor receptor 2 mediated cell signaling pathway is an important therapeutic strategy for HER2-positive cancers. Although monoclonal antibodies are currently used as marketed drugs, their large molecular weight, high cost of production and susceptibility to proteolysis could be a hurdle for long-term application. In this study, we reported a strategy for the development of artificial antibody based on
γ
-AApeptides to target HER2 extracellular domain (ECD). To achieve this, we synthesized a one-bead-two-compound (OBTC) library containing 320,000 cyclic
γ
-AApeptides, from which we identified a
γ
-AApeptide, M-3-6, that tightly binds to HER2 selectively. Subsequently, we designed an antibody-like dimer of M-3-6, named M-3-6-D, which showed excellent binding affinity toward HER2 comparable to monoclonal antibodies. Intriguingly, M-3-6-D was completely resistant toward enzymatic degradation. In addition, it could effectively inhibit the phosphorylation of HER2, as well as the downstream signaling pathways of AKT and ERK. Furthermore, M-3-6-D also efficiently inhibited cell proliferation
in vitro
, and suppressed tumor growth in SKBR3 xenograft model
in vivo
, implying its therapeutic potential for the treatment of cancers. Its small molecular weight, antibody-like property, resistance to proteolysis, may enable it a new generation of artificial antibody surrogate. Furthermore, our strategy of artificial antibody surrogate based on dimers of cyclic
γ
-AApeptides could be applied to a myriad of disease-related receptor targets in future.