The Agent Skills Directory

[COMMAND_EXECUTION]: The skill employs shell command templates in SKILL.md that utilize variable interpolation (e.g., $PKG). This design presents a potential command injection vector if the agent populates these variables with unsanitized input.- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it retrieves and processes external data from Google Play Store listings. Malicious instructions embedded in store descriptions or titles could attempt to manipulate the agent's behavior.
Ingestion points: Listing data is fetched using gps listings get and gps call gps_list_images in SKILL.md.
Boundary markers: There are no boundary markers or instructions to treat listing content as untrusted data.
Capability inventory: The skill possesses write capabilities including gps listings update and image modification tools.
Sanitization: The instructions do not define any sanitization or validation steps for content retrieved from external sources.

gps-listings