gallia-screenshot

by Kyle Bergstedt

Three capture layers let the AI see what it produces and self-correct without asking the user to screenshot manually. This is the core enabler for autonomous visual feedback loops.