C.26.1:4.5 - Well-formed probe-coupled boundary state
A probe-coupled boundary decision is usable only when the record states all of the following:
- the boundary, context bridge, service interface, team boundary, role relation, authority relation, or system edge involved;
- the probe lane and output carrier;
- the state reading before the interaction and the state reading after the interaction;
- the state-change evidence, including traces, changed labels, changed priorities, changed timings, changed routines, changed bridge fields, or changed downstream decisions;
- the local stop condition: which use the output does not support without another pattern;
- the neighboring FPF pattern that would carry the other claim.
The record is unfinished when any of these remains true:
- the output is named, but the operation that produced it is hidden;
- the operation is named, but the state that changed is not named;
- the state changed, but the decision that changes because of that fact is not named;
- the bridge/export loss is stated only as a vague warning rather than as a concrete non-admissible use;
- the same interaction is alternately treated as evidence, command, measurement, and bridge without role split.
The minimal admissible output is often enough: “this dashboard value is probe-coupled evidence for readiness behavior under window W”; “this workshop work changed alignment and therefore the workshop note cannot be treated as passive discovery”; “this API read is a non-neutral observation under these interface semantics”; “this context cut needs both F.9 bridge loss and C.26.1 probe-coupling treatment.”