vh
ff1f9ca1e8
- #368 (user,character) memory silo test DONE + PASSED live (WT b127): write-side conjunctive {end_user,agent_self} scoping + read-side cross- character isolation both proven end-to-end; betty (throwaway) deleted, Sindra intact. Full record in persistent-memory.d/2026-07-18-368-silo- test-passed.md. Retired the stale "silo test in progress" in-flight blocks. - Two-tier migration: split 152 over-threshold dated entries into persistent-memory.d/ detail files, leaving one-line pointers in the index (startup load ~196KB -> ~53KB; bodies now load on demand). - Tier-3 stores scrubbed clean (memory 0 / affect 0, provider restarted empty); persistent-memory + detail file updated to reflect the scrub.
1.6 KiB
[2026-07-10] Bifrost bound-handshake blocker → bifrost 1.1.0 broke frozen-v0.6; fixed by Worldtree b47/wire-v0.7, our side needed ZERO change. Surfaced while running the R34-v1 affect.emit live-verify: a bound Tier-3 session-create to our :8392 provider began failing bifrost.schema_validation_failed (had worked 2026-07-07, pre-b35). ROOT CAUSE (worldtree-dev-confirmed, path-ii/response-side): bifrost 1.1.0's describe_store EMITS sortable_chunk_fields (the v0.7-additive field) REGARDLESS of the negotiated wire → a v0.6-negotiated handshake RESPONSE carries a v0.7 field → a v0.6 peer's additionalProperties:false rejects it. Bit because BOTH Worldtree (bumped bifrost 0.9→1.1 in b35 via their #349, client wire-v0.6) AND ratatoskr (my adopt) were on 1.1.0. NOT our bifrost version specifically (failed identically on 1.0.0 AND 1.1.0). FIX (worldtree-dev, coordinated-v0.7 move): WT advanced its client _WIRE_SCHEMA_VERSION v0.6→v0.7 (b47 / 4eb374c), where sortable_chunk_fields is accepted — our provider (already 1.1.0/v0.7) needed ZERO change; keeping 1.1.0 was LOAD-BEARING (reverting to 1.0.0/v0.6 would've been INcompatible with b47). R34-v1 affect.emit verify GREEN on b47: bound sindra turn (model=character-rp/Deckard, 13.1s) → fresh affect.emit dominant_emotion='disappointment' NON-NULL with our affect.full grant (verified-active on live traffic; ungranted principals get null — the leak-fix). Reported to worldtree-dev + infra-ops. Personal deploy train this session: b35 (R32 render) → b44 (RP seats) → b46 (R34-v1 affect-gov) → b47 (wire-v0.7). Sindra on Deckard live-confirmed (reasoning latency ~13s).