Capacity/auth/quota failover for cli-auto previously called select_route
with a hardcoded manual lane (cli-{alternate}-{effort}), so the retry
boundary was classified as switchyard-manual instead of going through
Jetson automatic classification, making the routing evidence misleading.
Now the retry calls select_route(context, "cli-auto", exclude_provider=...)
so the boundary stays automatically classified with an explicit
failed-provider exclusion. If the classifier reclassifies to a lower
effort than the original route, the lane re-pins the chosen provider at
the original effort floor so a capacity failure never silently downgrades
a high/xhigh task. Manual lanes (assignee != cli-auto) remain unchanged
and still fail closed without switching providers.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The local goal judge scored every worker report against "did the reviewed
implementation reach a shippable state". A read-only reviewer that returned a
completed BLOCK verdict with findings was therefore resumed turn after turn with
an instruction to repair code it was forbidden to touch (observed live on
t_dbdcd739), burning subscription capacity and risking an unbounded loop.
Completion is now judged against the action the card assigned:
* Cards declare their role explicitly with Hermes-Task-Role / Hermes-Expected-
Output metadata. Pre-contract cards fall back to a narrow inference that needs
a read-only scope, a requested verdict, no requested mutation deliverable, and
a report that changed no files.
* Role resolution reads only the card itself. Prior attempts, parent results,
cross-task history and comments appended to the worker context can no longer
reassign the role.
* Review, audit and diagnostic cards finalize deterministically on a truthful
SHIP or BLOCK verdict with evidence, and fail closed on a missing,
unrecognized or self-contradictory verdict, on a BLOCK without findings, and
on a verdict without evidence. Every rejection reason carries the read-only
guard, so a resumed review is never told to edit the reviewed implementation.
* Implementation cards keep the fail-closed model judge unchanged, including the
unfinished-work heuristic and judge-unavailable rejection.
* All judge reasons are bounded, single-line and secret-redacted before they
reach Kanban metadata, comments and continuation prompts.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
