The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from existing presentations and replaces template placeholders, which represents a potential surface for indirect prompt injection if the source documents contain maliciously crafted instructions.
Ingestion points: The get_presentation action retrieves full details of slide content, and search_presentations returns metadata from the Google Slides environment.
Boundary markers: No specific delimiters or 'ignore embedded instructions' warnings are included for handling data returned from the Slides API.
Capability inventory: The skill possesses wide-ranging modification capabilities including batch_update, delete_object, and add_image (which fetches remote content), and it communicates with the vendor's official API.
Sanitization: The instructions do not define specific sanitization or escaping protocols for data retrieved from slide elements before it is processed by the agent.

google-slides