With increased reliance of digital storage for personal, financial, medical, and policy information, a greater demand for robust digital authentication and cybersecurity protection measures results. Current security options include alpha-numeric passwords, two factor authentication, and bio-metric options such as fingerprint or facial recognition. However, all of these methods are not without their drawbacks. This projects leverages the fact that the use of physical handwritten signatures is still prevalent in society, and the thoroughly trained process and motions of handwritten signatures is unique for every individual. Thus, a writing stylus that can authenticate its user via inertial signature detection is proposed, which classifies inertial measurement features for user identification. The current prototype consists of two triaxial accelerometers, one mounted at each of the stylus' terminal ends. Features extracted from how the pen is held, stroke styles, and writing speed can affect the stylus tip accelerations which leads to a unique signature detection and to deter forgery attacks. Novel, manual spatiotemporal features relating to such metrics were proposed and a multi-layer perceptron was utilized for binary classification. Results of a preliminary user study are promising with overall accuracy of 95.7%, sensitivity of 100%, and recall rate of 90%.