The Agent Skills Directory

[COMMAND_EXECUTION]: The skill interacts with the Google Drive API by executing the gws command-line binary. It supports a wide range of methods including files.create, files.delete, and permissions.update as part of its core functionality.- [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection as it is designed to read and process untrusted external data from Google Drive.
Ingestion points: Data is ingested through API methods such as files.get (with alt=media), files.download, comments.list, and replies.list as defined in SKILL.md.
Boundary markers: This file does not specify explicit boundary markers or 'ignore' instructions for the data it retrieves, although it references global security rules in ../gws-shared/SKILL.md.
Capability inventory: The skill has extensive capabilities to modify Drive resources, including the ability to change file permissions and delete revisions.
Sanitization: No input sanitization or validation protocols are described for the content retrieved from the Drive API within the provided instructions.

gws-drive