Nothing reports the proportion of intents that carry a mechanism claim alongside the entailment reading's findings

Nothing reports the proportion of intents that carry a mechanism claim alongside the entailment reading's findings

The readings companion (v4, section 6.6) bounds the entailment reading: the causal claim is prompted and nullable under the recording gradient, so where ## Mechanism is null the reading has nothing on the causal side to work from, and "the yield of this position depends on how many intents carry a mechanism. The proportion should be reported alongside the findings."

Nothing computes that proportion. The assembler's size report counts bytes and estimated tokens per material kind; the manifest enumerates items by path and field; the ingest verb validates what came back. None of them says how many of the projected intents carried a ## Mechanism section, how many carried the None stated. nullity, and how many carried neither. Today the answer for the cold-reading workstream's own fifteen shipped intents is zero, zero and fifteen, which is exactly the bound the companion wants stated beside the findings rather than discovered by a reader afterwards.

The natural home is the assembly result at the entailment position, where the intent projection already knows which fields each file contributed: a field a file does not carry contributes no item, so the count is already implicit in the manifest and only needs stating. The number belongs in the size report's rendering and in the manifest, so it is on the record for the run rather than recomputed at write-up.

Grounds

• pursued: we expect the yield bound the readings companion's section 6.6 asks for to be reported by the size report at assembly, so it is known before the run rather than counted after it; a report whose three counts did not sum to the projected intents it carried, or that stated the proportion at a position the companion does not ask it of, would show it wrong.