The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on shell commands executed via the gog utility to interact with Google Workspace APIs.
[EXTERNAL_DOWNLOADS]: The skill configuration automates the installation of the gogcli tool from the author's Homebrew tap (steipete/tap/gogcli). This is a resource owned by the skill author.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process untrusted external content from Google Workspace.
Ingestion points: Untrusted data enters the agent context through commands like gog gmail search, gog drive search, gog docs cat, and gog sheets get (SKILL.md).
Boundary markers: The instructions do not provide explicit boundary markers or warnings to ignore instructions embedded within retrieved emails or documents.
Capability inventory: The agent has the ability to perform sensitive actions based on processed data, such as sending emails (gog gmail send), creating calendar events (gog calendar create), and updating spreadsheets (gog sheets update).
Sanitization: No sanitization or validation of the retrieved content is described in the provided instructions.

gog