This paper presents the design and implementation of an educational workshop that integrates AI-based planning with robotic manipulation in a simulation environment. Developed as part of a Cognitive Robotics course, the workshop engages students in task and motion planning using the Unified Planning Framework and in executing their plans in Isaac Sim with a simulated Franka Emika robotic arm. The core learning activity involves manipulating oriented cubes using a robot manipulation language that abstracts gripper orientations, facilitating symbolic planning and spatial reasoning. Students complete a two-stage exercise: first, modeling and generating manipulation plans using UPF and PDDL; second, executing these plans in a high-fidelity simulation. The pedagogical design emphasizes integration of planning and execution and bridges high-level symbolic modeling and low-level robotic control, providing a rich, experiential learning environment. Findings suggest that such simulation-based activities enhance student understanding of AI planning and robotic affordances while making complex concepts accessible. The workshop offers a model for teaching cognitive robotics and highlights the potential of modern simulation tools in engineering education.