Reviewer set is not a hard dispatch invariant #156

New issue

Closed

opened 2026-08-15 14:33:52 +00:00 by john · 0 comments

john commented

2026-08-15 14:33:52 +00:00

Owner

Source: strict re-review C-12
Verdict: DISPUTE - subsumed, closing

Claim as reviewed

Reviewer-set resolution detects and falls back rather than establishing an invariant that an out-of-set reviewer cannot execute, so review independence can degrade rather than fail closed.

At HEAD `f354864` (v3.3.0)

Accurate. cadence-core/bin/route.mjs:630-636 emits reviewers on the resolve envelope as advice; the dispatch itself is performed by the orchestrating model following references/review-triggers.md. Nothing prevents a dispatch the resolve did not sanction.

Note

This is the same seam-advises / model-enforces shape as the risk_surface record issue, which is open and is the general form of the problem. Fixing observability there is what makes this one answerable.

Closed as a duplicate in substance; reopen if the risk_surface work does not generalize.

**Source:** strict re-review C-12 **Verdict:** DISPUTE - subsumed, closing ### Claim as reviewed Reviewer-set resolution detects and falls back rather than establishing an invariant that an out-of-set reviewer cannot execute, so review independence can degrade rather than fail closed. ### At HEAD f354864 (v3.3.0) Accurate. `cadence-core/bin/route.mjs:630-636` emits `reviewers` on the resolve envelope as advice; the dispatch itself is performed by the orchestrating model following `references/review-triggers.md`. Nothing prevents a dispatch the resolve did not sanction. ### Note This is the same seam-advises / model-enforces shape as the `risk_surface` record issue, which is open and is the general form of the problem. Fixing observability there is what makes this one answerable. Closed as a duplicate in substance; reopen if the risk_surface work does not generalize.