itd-5 preserves a fixed oracle cascade that the dependencies chapter says does not exist

itd-5 preserves a fixed oracle cascade that the dependencies chapter says does not exist

A premise contradiction (severity minor) found by the cross-document consistency pass.

• End A: .abcd/development/intents/disciplines/itd-5-prompt-quality-additions.md:87 — "Adding capability_scope does NOT modify the oracle cascade"
• End B: .abcd/development/brief/02-constraints/02-dependencies.md:35 — "There is no fixed RP→codex→in-session cascade."

itd-5 says adding capability_scope does NOT modify the oracle cascade, and designs capability routing as a selector layered above that cascade. The dependencies chapter says the oracle is host-delegated by default with opt-in adapters and that there is no fixed RP→codex→in-session cascade. The discipline's design rests on a mechanism the current architecture has removed.

Evidence: .abcd/work/reviews/2026-10-26-consistency/00-summary.md, finding 144.