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.
1.1 KiB
[2026-07-15] soong-lab bundle ship.native.motivational.goals/fears are bare STRINGS, but WT agents.define requires OBJECTS {id, type∈{maintenance,achievement,avoidance}, salience[0,1], description≥20ch} (ValidatedGoal/ValidatedFear, worldtree core/conversation_api/api.py:2116, extra=forbid; ids unique across goals+fears). So ship.native is NOT directly define-valid on motivational (soong-lab's Frame Invariant 1 breaks there). A bundle→WT consumer MUST map string→object + synthesize id/type/salience. Informed soong-dev → RESOLVED soong-lab v0.3.24 (2026-07-16): now emits WT-valid objects (type+salience captured, id synthesized at export, validate_exportable gates description≥20/enum/range) → drop the string→object workaround for v0.3.24+ bundles; legacy pre-fix designs coerce on open, so re-exports are valid too. Rule: PROBE a define payload under a throwaway agent_name BEFORE a destructive DELETE+redefine — validating via sindra-probe2→201 caught the 422 without leaving the real Sindra deleted+undefined. (A failed probe still RESERVES the name → 409 on retry; use a fresh probe name.)