lore

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s sync/compress workflows read user-edited project files under .lore/ (e.g., existing ## My notes seed during init step 5 and entry text during compress step 1/3) which can contain outsider-authored free text in the repository and is consumed by the agent at runtime.