This work presents the development and implementation of an interface to control a robotic exoskeleton for gait rehabilitation in patients with stroke through exercises based on conventional neuro-rehabilitation therapies. Movement patterns of each exercise were acquired using an optoelectronic ten-camera system in a group of subjects without pathology. An interface for robotic control was developed in the LABVIEW environment that allows kinematics of each exercise to be reproduced in a robotic exoskeleton. The interface allows us to control: the assistance percentage, movement patterns, movement speed, number of repetitions, and resting time. The software presents plot panels for each joint that shows programmed kinematic and executed curves in real-time as a feedback tool for therapist. Finally, the kinematic data is stored in a directory with the patient’s name and session date for post-analysis.