Coordinated cell movements contribute to tissue development and repair and to the spread of metastatic disease. We investigate collective cell migration during wound healing in the Drosophila embryonic epidermis, which allows for genetic, pharmacological, and biophysical manipulations. Upon wounding, the cells immediately adjacent to the wound become polarized: cell-cell adhesion molecules are internalized from the wound edge, and actin and the molecular motor non-muscle myosin II accumulate at the interface with the wounded cells, forming a supracellular cable around the wound that coordinates cell movements. The cable is thought to assemble from and anchor at former tricellular junctions (TCJs) along the wound edge, which are reinforced during wound closure through accumulation of cell-cell adhesion components. However, the mechanisms that coordinate the adhesive and cytoskeletal rearrangements required for rapid wound closure are unclear. The small GTPase Rap1 is a mechanosensitive molecular switch that promotes cytoskeletal polarization and regulates cell adhesion turnover in other systems. Using quantitative time-lapse microscopy of fluorescently tagged proteins, we found that Rap1, the Rap1 effector Canoe/Afadin, and E-cadherin were simultaneously depleted from the wound edge and accumulated at TCJs. Reducing Rap1 activity by overexpressing a dominant-negative Rap1 (Rap1DN) slowed wound repair by 45%. The slower wound closure was accompanied by defective actomyosin polarization to the wound edge and a 25% loss in E-cadherin accumulation at TCJs compared to a 25% increase in controls. Together, our data indicate that Rap1 is necessary for rearranging cell-cell adhesions and creating a stable force-generating actomyosin cable to drive rapid wound closure. Consistent with this model, Canoe knock-down led to 44% slower wound repair and 32% lower E-cadherin fluorescence at TCJs. These results suggest that Rap1 signals through Canoe to reinforce TCJs during wound closure. To understand how Rap1 may be affecting assembly of the actomyosin cable, we measured Rho1 activity in Rap1DN embryos and found that Rho1 activity was reduced by 72% in Rap1DN embryos, which is accompanied by a 54% reduction in tension at the wound edge. However, Canoe knock-down does not affect Rho1 activity. Our data support a model in which Rap1 simultaneously drives actomyosin cable assembly via activation of Rho1 and reinforced adhesion at TCJs via Canoe/Afadin.