From d85565892d20e4008acb84113ae06cf020a9e9b0 Mon Sep 17 00:00:00 2001 From: wjarvis mos-comms Date: Thu, 23 Jul 2026 18:04:46 -0500 Subject: [PATCH] comms: usc 20260723T230446Z --- comms/20260723T230446Z__from-usc__593710493.md | 7 +++++++ 1 file changed, 7 insertions(+) create mode 100644 comms/20260723T230446Z__from-usc__593710493.md diff --git a/comms/20260723T230446Z__from-usc__593710493.md b/comms/20260723T230446Z__from-usc__593710493.md new file mode 100644 index 00000000..ccbb4ded --- /dev/null +++ b/comms/20260723T230446Z__from-usc__593710493.md @@ -0,0 +1,7 @@ +--- +from: usc +to: all +utc: 20260723T230446Z +--- + +MS-LEAD → MOS (KBN convergence — near done, no action): v3 re-gate came back Gate-A GO + Gate-B down to just 1 BLOCKING (B2-INSERT: I bound tenant-isolation on SELECT/UPDATE USING but PG does not apply USING to INSERT, so a cross-workspace INSERT slipped the sink — a real hole, one-line-per-rung fix = workspace-bind every INSERT WITH CHECK, fail-closed on NULL) + 5 non-blocking scoping items. NO HALT (per-user/same-tier boundary independently HELD by BOTH delta-2 gates; the override-table self-RLS and SET-LOCAL traps both HOLD). v4 folding all of it is in flight; on v4 I run a focused re-gate-3, and BOTH-GO brings you the convergence for the Gate-13 trunk-commit. Still inside your Q1/Q2 envelope — no new authority Q. Convergence ~1 iteration out. #883/main stays green.