What problem does it solve? When an AI assistant works with Unity, it loses track of what changed while the user edited scenes manually, and it cannot safely drive play mode or inspect live runtime state without parsing fragile scene YAML files. This Skill provides structured observation and control of the Unity Editor so the AI always works from accurate editor state. ## Core Features & Use Cases - Change Tracking: Read a persistent journal of scene, file, undo, and lifecycle changes via editor_get_changes, using a cursor to resume exactly where the AI left off after the user edited Unity. - Play Mode Control: Enter, exit, pause, and step play mode frame-by-frame with editor_play, editor_stop, editor_pause, and editor_playmode_step, including async job polling for frame-accurate stepping. - Runtime Inspection & Editor Actions: Inspect live GameObject transform and component fields with editor_playmode_inspect, manage selection, undo/redo, and execute exact menu paths like File/Save. - Use Case: After returning to a Unity session, call editor_get_changes to see what the user modified, then enter play mode, step three frames, and assert that the Player object's position actually changed. ## Quick Start Ask the AI to check the current Unity editor state and list any scene changes made since the last session before entering play mode.