Cold Review

Cold review

A cold reviewer is not an extra panelist. It is a control on the panel.

A panel converges. Reviewers read each other, findings get confirmed by seats already looking in the same direction, and the group arrives somewhere with more confidence than any member earned alone. That convergence is usually right, which is why it is dangerous when it is wrong: a panel of three has approved a head carrying a defect all three missed, and what surfaced it was a differently framed read, not a fourth verifier.

This file is read by the person who WRITES the brief, not by the seat being briefed. The value is in the brief, not in the seat's context. A cold seat that never loaded a skill file, never joined the mesh and received nothing but a persona has produced three blockers that two fully-briefed lenses missed. So nothing here requires the graded seat to have access to this document, and no rule below may be written so that it does.

The isolation

• Never joins the review channel. Not muted, not quiet. Not joined.
• Never shown the panel's findings, verdict, or round count, including as a list of things already ruled out so it can skip them. Skipping-instructions are findings shaped like scope.
• DM-only or file-only input from the author or manager. Its verdict goes first-hand to the dedicated record destination itself.
• Never shares a model family with whoever WROTE the change.

Why never join, stated correctly. Joining exposes the seat to the panel's live traffic for as long as it is subscribed, and the panel is active during a cold read. Replay of history is a second exposure path but not a guaranteed one: replay is channel.replay ?? defaults.replay ?? true, a default-true policy with a per-channel override, so a replay=false channel would not replay. The rule does not depend on that setting in either direction, and a brief that justifies it by replay alone is resting on a config it cannot rely on.

Prefer the ACL to the request. subscribe: [] plus an allowSubscribe that excludes the panel channel makes non-join a property of the seat rather than an instruction it must follow. Where a fence exists, use the fence.

If you catch yourself wanting to tell it something so it does not waste effort, that is the anchor forming. Let it waste the effort; that is the price of the control, and it is cheap next to a laundered verdict.

What the brief contains

It must be executable without a follow-up question, because a follow-up is the contact the isolation forbids. A brief that omits the operational half forces the seat to come back and ask.

The substance, and nothing beyond it:

• the artifact at a named exact commit, and the instruction to re-resolve that sha itself;
• the bounded question, stated so it cannot be answered by agreeing with someone;
• for a delta re-grade, the exact sha range and an explicit scope boundary naming what is out of scope and already settled.

The operations, all of which are required and none of which anchor:

• the absolute path of its own detached worktree, which is never the graded tree;
• where the verdict goes, by exact destination (see below);
• the prohibitions it operates under, and the verdict shape;
• its own model and effort line, to be stated in the verdict.

What must never appear: the issue's own diagnosis, the author's rationale, "the tricky part is X", the panel's findings in any form, or a question phrased as confirmation. Ask "what does this change accept that it should not, and what does it reject that it should not" and let the seat derive the set. Never "do you agree that".

If the author believes something is weak, that belief goes to the panel, which benefits from it. The cold seat's job is to find what nobody framed.

Where the verdict goes

A verdict that exists only in a manager's context is not a verdict. It must land somewhere that survives the seat and the manager, and the seat must put it there itself, because a relayed verdict cannot be distinguished from an invented one.

The broker-attested destination is a Cotal mesh post on one dedicated record channel that the cold seat may publish only to and the panel cannot read. The cold persona keeps subscribe: [] and allowSubscribe: [], and grants allowPublish only for that record channel. Publishing the verdict does not require joining it.

A file at an absolute path written into the seat's persona remains an explicit fallback when the record-channel route is unavailable. It is first-hand delivery only as a norm and cannot satisfy a gate that requires broker-attested poster identity. A GitHub comment is not a substitute: it identifies the workstation's GitHub account, not the Cotal seat, and a shared gh credential lets the manager post the same comment.

Poster is a control only on the Cotal broker-attested destination. A mesh post carries the Cotal principal the broker records. A GitHub comment carries a GitHub account and does not bind that account to the Cotal seat. A file at an absolute path carries neither: any process that can write that path can produce the artifact, including the manager who later "confirms" it by re-reading it. First-hand file delivery is therefore a norm on the seat and the briefer, not a control a later stranger can verify. A gate that treats the file's existence as proof the seat posted has accepted a relay. When the brief must use a file, say so in the verdict record, and do not list poster or channel membership among the checkable controls for that delivery.

Verify at the destination, never at the source. Re-fetch the landed artifact and grep it for content you expect, with a positive control so an empty fetch cannot pass as a clean result. Writes that report success and do not land are common, including an edit that reports nothing and changes nothing or a file write that does not persist while the tree reports clean.

What the verdict says

APPROVE, or named blockers. Never an open-ended re-read, never "looks fine so far", never a list of things it might check next.

It must separate what it EXERCISED from what it INSPECTED, and the split is defined by the OBJECT, not the verb: exercised means the reviewed artifact itself ran through a real entry point. Running a tool to read the artifact, testing against a mock, compiling, or a dry run are inspection. A verb list cannot settle those cases and two seats will classify them differently.

Where it could not exercise something, it names the gap and why. A named gap is a limit of the ENVIRONMENT, not a licence for a limit of EFFORT. "I graded this by reading because running it would take the fleet down" is a boundary on what is knowable; "I did not get to that part" is a boundary on what was attempted, and the two must never be written in the same words.

A blocking finding must enumerate the attacks that FAILED. Without the survey, one success reads as a lucky hit; with it, the finding is a surveyed surface with one hole, and the fixer learns which ground is already covered.

What a cold verdict binds, and who may override it

A cold verdict is not a veto. It is binding as a question that must be answered at the artifact, publicly.

This matters because isolation defends against panel anchoring and not against error anchoring. If the panel disproved claim C with evidence E, the cold seat may re-derive C and report it with fresh confidence, and every route back to it is forbidden: showing E is showing findings, saying "C is settled" is a skipping-instruction, and asking it to recheck is confirmation framing.

The resolution never required talking to the seat:

• A manager may override only if it neither authored the change nor graded or supervised the panel. It verifies the claim against the artifact itself and publishes the refutation with its own evidence, under its own name, leaving the cold finding standing in the record beside it. Nothing flows back to the seat, so isolation holds; the override is auditable, so the control survives. The forbidden act is a PRIVATE override, not an override.
• An author may not override a cold verdict on their own change. They may fold it, or request a fresh permitted stand-in, who overrides publicly under their own name. Publication separates override from launder only when the publisher and the accused are different parties: an author refuting a finding about their own work has produced a document, not a check.
• A permitted stand-in is an independence test, not a signature. It had no part in authoring, grading or supervising the change, never shares a model family with the author, and verifies the claim at the artifact without the author's account of it as input. A coordinator may provision a fresh referee in an isolated worktree and give it only the artifact, the finding and the exact sha; whoever provisions the referee does not thereby become the referee. A prior panelist, lane manager or supervising coordinator is not a stand-in. A clean byline on the author's reasoning is the original hole with a different name.
• Every public refutation names the exact sha it answers and closes the blocker only at that sha. A moved head requires a new verification and a new public record; a refutation is never carried across a sha any more than a verdict is.
• A cold blocker is closed by the seat's own APPROVE or by a permitted public refutation, and by nothing else. Any completion gate must first require a landed terminal verdict, APPROVE or named blockers, before it evaluates closure. A silent, failed, or non-verdict lifecycle cannot pass on an empty finding set. After that prerequisite, the gate must accept both closure routes. A gate that demands the seat's approval alone deadlocks the override the first time it is used correctly, because the seat is one-delivery and may never be told to reconsider: the only exits left are satisfying a finding that was just publicly refuted, or a zero-delta re-pin to manufacture an approval, which is laundering.