What this core is for
Background service
Docs
Background memory extraction layer that scans chat history, stores user and room memory, and feeds Hydra context.
Background service
Memory Core is background infrastructure. It scans chat history, extracts durable facts, and injects memory context back into Hydra instead of acting like a direct Verba surface.
Max completion tokens for memory extraction.
Key: extraction_max_tokens Default: 2700Max memory facts injected per scope (user and room).
Key: hydra_max_items Default: 12Max characters for user/room memory summaries in context.
Key: hydra_summary_max_chars Default: 2100Max characters per memory fact value in context.
Key: hydra_value_max_chars Default: 288How often memory extraction runs.
Key: interval_seconds Default: 180 Options: 1 minute, 2 minutes, 3 minutes, 5 minutes, 10 minutes, 15 minutes, 30 minutes, 1 hourMax new messages processed per scope each pass.
Key: lookback_limit Default: 80Minimum confidence required before storing facts.
Key: min_confidence Default: 0.65Use Settings > People links to merge selected portal users into one master memory profile.
Key: use_people_identities Default: TrueEnable writes to mem:room:{platform}:{room_id}.
Key: write_room_memory Default: TrueEnable writes to durable user docs. Linked People use mem:user:identity:{person_id}; unlinked portal users use mem:user:{platform}:{user_id}.
Key: write_user_memory Default: True