We introduce SkyNetz, a playful interactive robotics simulator for computer science students. Its focus lies on the visualization of a probabilistic robot localization algorithm considering noise in the sensors and actuators of the robot on the one hand, as well as an environment filled with obstacles which damage the robot on contact on the other hand. The goal of the simulation is training students on the intricacies of the algorithm and to develop a notion on the impact of the considered factors such as the degree of sustained sensory noise. In order to facilitate learning and promote exploration, we embed a game mode that conveys the basic interactions with the simulator and the factors shaping the robot's behavior. In the game, the player helps the simulated robot to reach its destination with as little damage as possible. This is done by setting waypoints for the robot by adjusting the parameters of the deployed localization algorithm as well as the quality of sensors and the accuracy of the robots movements. By playing with these parameters, the user playfully learns their effects, which are visualized in 3D - contrary to the hard mathematical approach presented in books. SkyNetz also has the capacity to communicate with a real robot, to show its current position and position estimates. In the long run, this will provide the foundation for novel augmented reality games. The paper includes a general introduction to the topic of interactive robot simulation, background on the specific problem of localization estimation, the presentation of our approach and the results from a small user study.