Spasticity is an important factor in designing wearable and lightweight exoskeleton neurorehabilitation robots. The proposed study evaluates biomechanical reactions of an exoskeleton robot to spasticity and establishes relevant guidelines for robot design. A two-axis exoskeleton robot is used to evaluate a group of 20 patients post-stroke with spastic elbow and/or wrist joints. All subjects are given isokinetic movements at various angular velocities within the capable range of motion for both joints. The resistance torque and corresponding angular position at each joint are recorded continuously. Maximal resistance torques caused by low (modified Ashworth scale (MAS) 0, 1), intermediate (MAS 1+), and high (MAS 2 and 3) grade spasticity were 3.68 ± 2.42, 5.94 ± 2.55, and 8.25 ± 3.35 Nm for the elbow flexor ( , between each grades) and 4.23 ± 1.75, 5.68 ± 1.96, and 5.44 ± 2.02 Nm for the wrist flexor ( , for low versus intermediate, low versus high grade spasticity). The angular velocity did not significantly influence maximal resistance torque in either joint. The catch occurred more quickly at higher velocities for low and intermediate elbow flexor spasticity ( ). Spasticity caused considerable resistance to the robots during mechanically actuated movements. The resistance range according to the degree of spasticity should be considered when designing practical neurorehabilitation robots.