The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface through the ingestion of untrusted data.
Ingestion points: The {{ARGUMENTS}} placeholder in SKILL.md ingests arbitrary user text.
Boundary markers: Absent; no delimiters isolate the user input from the rest of the prompt context.
Capability inventory: The skill triggers shell execution via the omc ask command.
Sanitization: Absent; no input cleaning or escaping mechanisms are described.
[COMMAND_EXECUTION]: User-supplied text is interpolated directly into the shell command template omc ask {{ARGUMENTS}}, which could allow command injection if arguments are not safely handled by the agent platform.
[EXTERNAL_DOWNLOADS]: The documentation guides the installation of the agy binary from the official Google domain https://antigravity.google.

ask