The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional Markdown files without executable code, scripts, or external network operations. The content focus is on strategic planning and operational workflows.
[DATA_EXPOSURE_AND_EXFILTRATION]: The instructions explicitly mandate privacy best practices, such as separating sensitive attendee logistics (medical, dietary, travel) from admission decisions and requiring explicit consent for public profile sharing and model training.
[INDIRECT_PROMPT_INJECTION]: The skill defines a process for the agent to ingest external data (e.g., "customer evidence," "launch brief," "attendee links"). While this represents a standard injection surface, the skill includes instructions for the agent to let users review and approve recommendations before they are finalized, serving as a mitigation.

gtm-plan