update-brain

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). In SKILL.md, update-brain’s required runtime workflow reads only user-specific cursor-windowed content from connected tools tagged context (and drains its own crawl_frontier queue), rather than ingesting arbitrary outsider-authored free text without an explicit selection/search-from-tools step; Supermemory memory/recall are internal fact storage, not an outsider text source.