Voice-regression watch: a general token guard

Make the voice-prompt slot-coverage drift guard collect every {{token}} from every watched agent's prompt AND config JSON and fail when one is missing from DYNAMIC_VAR_DEFAULTS, add the two missing defaults, retire the subsumed name pins, and unblind the three maintenance watch scenarios.

Status is DERIVED from each row’s own PR, decision and prod-evidence links by bin/refresh-tracker voice-watch-token-guard — never typed. Last refreshed 2026-10-22 22:36Z.

Overall
—
 
—
Shipped
—
Merged
—
In flight
—
Open
—
Waiting on you
Groups — click to filter
Status
—

Every row opens: when it was added, why it exists, and the links that prove it. Rows waiting on a human carry a Decide → link straight to the answerable question on the decisions page. Statuses are re-derived from live PR, decision and prod-evidence state by bin/refresh-trackers; nothing here is typed by hand. Merged and shipped both count as done (Gera, 2026-10-19); a prod_evidence permalink is what moves a row from Merged to Shipped.

PropFlow Docs