An acceptance criterion whose producer does not exist is judged by whether this phase wires it, not by whether the narrowing was disclosed

The rule

A fidelity audit regularly meets a criterion whose Given clause presupposes something no code can produce: the constructor refuses the value, or the verb that would mint it belongs to an intent that has not landed. One question decides the verdict, and it is not whether the spec admitted the gap.

Does this phase wire it?

• Wired in this phase — the criterion is MET_WITH_CONCERNS, and the verdict names the intent that closes it.
• Not wired in this phase — the criterion is NOT_MET, and the verdict carries the reason a promised intent goes unwired.

Disclosure is not the test. A spec that states its own sequencing up front is still a spec whose promise is unkept if nothing in the phase closes it, and an auditor that accepts disclosure as satisfaction grades the writing rather than the delivery.

The second half of the rule is the point of the first. NOT_MET obliges somebody to answer why a promised intent is not being wired — and that question, asked at the audit rather than at the retrospective, is what the verdict exists to force.

Dictated by the maintainer on 2026-08-30 and formatted; the ruling arose from two verdicts that turned on it.

The gate

• Given a criterion whose producer is absent from the delivered code, when the auditor judges it, then the verdict turns on whether an intent in this phase wires it, and names that intent.
• Given a criterion recorded MET_WITH_CONCERNS on the strength of a scheduled intent, when the phase closes, then the verdict is re-checked, and flips to NOT_MET if that intent did not land.
• Given a criterion judged NOT_MET for want of a producer, when its verdict is read, then the record states why the promised intent is unwired.

Fit

The discipline family is the right home: this fixes how a judgement is made across every audit, and has no user moment of its own. It sits beside the itd-1 acceptance-gate discipline and is consumed by the intent-auditor's definition rather than by a verb.

A stance sits inside this rule — a promise is judged against the phase, not against its own disclosure — which could stand as a principle. It stays here instead: one rule, one record, and the discipline carries the stance rather than duplicating it. If the re-check at phase close proves to need machinery, that is the point at which this graduates to an ADR.

Staging

The rule is applied by hand today: two criteria in the current cycle were judged under it, each recorded MET_WITH_CONCERNS against a scheduled intent, each carrying the flip condition in the decision record.

Wiring is the unbuilt half, and it is two pieces: the judging rule stated in the intent-auditor's definition so an auditor applies it without being told, and a re-check at phase close so a conditional verdict cannot quietly survive the intent it depended on. Both wait on a plan; this record is the stub that holds the rule until then.