Obtaining and maintaining large dataset of network traffic patterns is necessary requirement for many types of research. The goal of this thesis is to automate the process of collection of cellphone network traffic. Because of the ever changing interface of current mobile applications, the tool will use mechanical control of the selected application (computer controlled robotic arm). The collection process will be controlled by playbooks (recorded sequence of user actions) and will allow the user to adjust the controls to new version of given application. The tool will be demonstrated on collection of voice call network traffic from selected application.