Found while proving CAROL-INI-3873 (b) on a live turn, and PRE-EXISTING: the same probe refuses identically on the pre-change code, so this is not caused by the engine convergence\u2026

Ninad ruling, 2026-08-16 (CLI-268), given when CAROL-INI-3868 slice (b) hit it head-on: "Agent's memory is user specific, period. If the same user appears on internal chat and ext\u2026

Ninad ruling (2026-08-15, CLI-262): a user talking to an agent gets the SAME experience irrespective of channel - WhatsApp or web. Recorded as a DESIGN POLICY and implemented unif\u2026

Ninad ruling 2026-08-12 (CLI-245): the Agent Chat service (id whatsapp) holds ONLY the three per-agent chat tracks. Move agent-resources-leo/carol/sam-chat into it (keys never cha\u2026

Ninad ruling (CLI-243): a user is one PERSON; the channels a person reaches Carol on live in a separate user-channel mapping. Merge the two Ninad rows (WhatsApp admin + web ninad1\u2026

Ninad ruling (CLI-243): start the user list clean. Keep Ninad (WhatsApp admin + web [email protected], access admin, carolverse role contributor), Werner and Sachin (promoted to\u2026

Ninad rulings 2026-08-11 after a live stranger-mode test double-minted a user and the turn was refused on the sales track: (1) an anonymous session creates EXACTLY ONE user record\u2026

Carols request classifier (5-min cron) re-asks the model about the same conversation forever: a parse failure writes NOTHING to the row, and an unsure verdict keeps the row in the\u2026

SURFACED BY CAROL-INI-3679 (CLI-229, 2026-08-04) while curing Carol's WhatsApp lane. MEASURED, all 70 pins in the live LLM config, none sampled. Each was classified by asking the\u2026