This study developed an active and flexible hand prosthetic model, inspired by the skeletal structure of the human hand, and controlled by electroencephalographic (EEG) signals. Computer-aided design (CAD) and 3D printing techniques enabled the integration of structural and muscular elements of the hand. An EEG-based control system was implemented to classify hand movements using a convolutional neural network (CNN). The CNN model achieved a 92% classification accuracy during real-time assessments of hand movements. The prosthesis exhibited a 95% success rate in adapting to various object shapes and sizes, with users reporting an average satisfaction score of 8.7 out of 10 for ease of use and functionality. The study utilized a primary dataset of 5000 EEG recordings from 50 participants, ensuring robust and diverse data for training and testing.