Control-moment gyroscopes (CMGs) are power-efficient, internal momentum actuators that produce high torque for the attitude control of spacecraft. CMGs are proposed for actuating joint degrees of freedom in a spacecraft-mounted, agile robotic payload. A kinematics and dynamics analysis is performed for a general open-chain, N -link, N -degree-of-freedom robotic system actuated by CMGs. For an example open-loop maneuver, a CMG system is compared to a system driven by reaction-wheel assemblies (RWAs), which are alternative internal momentum actuators. Numerical simulations demonstrate that a CMG system offers the same agility while using less than 1% of the power of a RWA system with identical dynamics and mass properties. A statistical study demonstrates that only CMGs can provide the output torque necessary to meet the agility requirements of the slew. With the established kinematics and dynamics for a CMG robotic system, numerical simulations are performed for a general CMG system manipulating a payload. The analysis of an added payload's effects on otherwise reactionless CMG systems motivates the exploration of possible operations concepts for reducing base reactions and power consumption. Simulation results for an example closed-loop maneuver show that base reactions can be significantly reduced, or even eliminated, with CMG actuation while using the same amount of power as a robotic system driven by conventional joint motors. Power-optimal steering is investigated for a CMG telescope application. A real-time optimization method is presented that includes null motion in a closedloop end-e?ector tracking problem. For a redundant robotic system, there are an infinite number of joint-angle solutions corresponding to a given end-effector attitude. In this optimization algorithm, the joint-angle command corresponding to a commanded end-effector attitude is adjusted with a null-angle component to minimize power while the tracking accuracy remains unchanged. Calculation of the null-angle component is based on a quadratic cost function, which is the sum of the squares of power for each CMG gimbal. Simulation results for an example maneuver demonstrate that the power consumption of the system is reduced by up to 38% when null motion is included in the feedback loop.