The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external Zoom meeting details, participant lists, and AI-generated summaries, which creates an attack surface for indirect instructions.
Ingestion points: Data is fetched via tools such as ZOOM_GET_A_MEETING, ZOOM_GET_PAST_MEETING_PARTICIPANTS, ZOOM_LIST_ALL_RECORDINGS, and ZOOM_GET_A_MEETING_SUMMARY from external Zoom records.
Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between meeting metadata (e.g., topics, participant names) and its core instructions.
Capability inventory: The skill has capabilities to create, update, and delete meetings, as well as manage recordings and participants.
Sanitization: The instructions do not specify any validation or sanitization of the content retrieved from the Zoom API before it is processed by the agent.

zoom-automation