memory-store

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow can ingest outsider-authored free text via the /diary → /reflect pipeline: raw session transcripts in diary/ (which may include messages/comments from other people) are analyzed to extract patterns and then synced into CLAUDE.md, which is fed back into the agent’s LLM context.