A numerical predictor-corrector (NPC) method for trajectory planning and closed-loop guidance of low lift-to-drag (L/D) ratio vehicles during the skip entry phase of a lunar-return mission is presented. The strategy calls for controlling the trajectory by modulation of the magnitude of the vehicle's bank angle. The magnitude of the bank angle used in the skip phase is determined by satisfying the downrange requirement to the landing site. The problem is formulated as a nonlinear univariate root-finding problem. Full three degree of freedom (3DOF) nonlinear trajectory dynamics are included to achieve high accuracy of the landing prediction. In addition, the proposed approach automatically yields a direct entry trajectory when the downrange is such that a skip entry is no longer necessary. The same algorithm repeatedly applied on-board in every guidance cycle realizes closed-loop guidance in the skip entry phase. A number of issues are identified and addressed that are critical in closed-loop implementations. Extensive 3DOF dispersion simulations are performed to evaluate the performance of the proposed approach, and the results demonstrate very reliable and robust performance of the algorithm in highly stressful dispersed conditions. Comparison is made between the proposed algorithm and an earlier skip algorithm developed for the Apollo space program. It is shown that the proposed algorithm is superior to the Apollo algorithm especially when used for entries with long downranges.