share-memory

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

• Third-party content exposure detected (high risk: 0.75). Runtime path: the skill’s required workflow reads and injects project-authored Markdown from AI_MEMORY/ (e.g., AI_MEMORY/PROJECT.md Long-Term Memory and AI_MEMORY/SYNC_LOG.md daily blocks) into the agent’s LLM context; those files may contain outsider-authored free text (e.g., from other agents or prior runs) and are not sanitized against prompt-injection.