This study investigates the measurement reliability of a stereo camera system designed to capture hand movements without the need for external markers or wearable devices. Such a markerless approach is especially beneficial for children with limited mobility or hand injuries, as it allows the assessment of fine motor actions in a completely non-intrusive manner. To examine the performance of the system, standardized hand movement exercises were recorded and analyzed over multiple repetitions. The evaluation focuses on the precision and consistency of the system, revealing the operational conditions under which the stereo camera provides reliable motion data. The findings support the development of non-invasive assessment methods and offer guidance for establishing robust markerless, automated hand motion analysis protocols.