Human-Robot Interaction (HRI) often suffers from a lack of transparency, making it difficult for users to interpret robot behaviours, anticipate movements, and feel secure during interactions. This paper presents a technical framework that integrates Extended Reality (XR) with robotics to enhance user understanding through real-time visualisation. Our primary contribution is the development of an integration pipeline that connects XR headsets with the TIAGo robot via the Robot Operating System (ROS). Leveraging recent advancements in XR hardware, including improved spatial mapping and high-resolution passthrough, the system accurately transforms coordinate data between the XR environment and the robot’s localised map. This enables real-time visualisation of critical information such as the robot’s planned trajectory and pose. A pilot study was then conducted to validate this system architecture as a tool for HRI research, demonstrating that augmented visual cues can significantly improve spatial awareness, trust and user intuition. The findings support XR’s role in fostering safer and more intuitive HRI. This research lays the foundation for future developments in XR-driven robotic interaction. The associated repository with the code, setup guide and pilot study data are available at: https://github.com/LCAS/XRVis_for_robots