max / quasi

A region can say its children are answers to one question Decided in quasicoherent f7abbc08 (Max chose c): a region member, not a rework of Repeat. Repeat describes a repeating field -- an Instance is one value and one error -- so a slot is one answer to one question. Nothing described a repeating group, and audiofiles' rule editor has two on one screen: a condition is three questions that only mean anything together, and an action is two. The regions already existed, 23 Group sites in the tree, so this attaches to something with consumers. Slot::repeating carries the singular noun, the floor, the ceiling and the add control. Slot::removes goes on each child, which is the one place this diverges from the task's sketch of "the add and remove actions" on the member: a parent-level remove has to reach a particular slot, and both ways to do that are worse -- a renderer building `.../{at}/remove` is route construction in a renderer, and one route plus an index under an agreed name is a third word beside SELECTED and TICKED for one screen. A child carrying the action that concerns it is fed_by's shape, already here. What the floor buys is the enforcement. audiofiles disabled its own last Remove, which put the rule in an app where every renderer needs it; Repeating::least says it once and all three read may_remove. The boundary is drawn dead rather than hidden, which is the call that editor had already made. Every renderer derives the chrome as nodes and draws it through its own walk, so none of them invents markup or styling: a slot's number is the same heading a described one gets and both controls are Node::Act. The number is the renderer's, one-based, because a description that wrote numbers goes stale the moment a slot leaves the middle. quasi-tui's drawing, measurement and focus walk read one pair of helpers so the three cannot come apart. No per-slot error, which is the one thing option (a) had. That is written down as the reopening condition rather than left implicit: a screen wanting an error against one slot of a repeating group is the evidence that the wider Repeat was the answer.