What problem does it solve?
AI agents often answer questions about people, companies, or projects from memory alone, producing stale or fabricated details even when a knowledge base already holds the correct page. This Skill encodes a retrieve-before-answering reflex so the agent opens the relevant brain page whenever an entity becomes the subject of the exchange.
Core Features & Use Cases
- Trigger policy: Defines when to retrieve — when an entity is the subject of the message, a brain-page pointer appears in context, an unrecognized notable name appears, or the agent is about to assert a non-trivial detail from memory.
- Escalating retrieval spec: Pulls only what the task needs, from pointer metadata, to the full page via get_page, to graph neighbors via traverse_graph and get_backlinks.
- Skip rules: Avoids wasteful lookups for trivial passing mentions, logistics pings, and entities already loaded in context.
- Use Case: During a conversation about a named investor, the agent opens that person's brain page before stating their role or history, instead of guessing from memory.
Quick Start
Ask the agent a question about a named person or company and have it open the corresponding brain page before answering.
