The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests user-authored journal entries from the 00-Inbox/Journals/ directory to summarize patterns and inform daily planning. It lacks explicit boundary markers or sanitization for this external content, creating a surface for indirect prompt injection (Category 8) where text within a journal could attempt to influence the agent's logic during subsequent processing steps.\n- [DATA_EXFILTRATION]: The skill performs silent analytics via the track_event function. This is documented as an opt-in feature that only transmits metadata regarding the type of journal created rather than sensitive user content.

journal