The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to use the gplay CLI tool, involving several shell commands for pulling, validating, and applying metadata changes to the Google Play Store.
[INDIRECT_PROMPT_INJECTION]: The skill enables a workflow where external, untrusted content from the Play Store is ingested into the agent's working environment.
Ingestion points: The metadata pull and images pull commands (referenced in SKILL.md) fetch live listing text and image metadata directly from the Google Play Store.
Boundary markers: There are no specified boundary markers or isolation instructions to prevent the agent from processing or obeying instructions that might be embedded within the listing text.
Capability inventory: The gplay CLI tool used by the skill has capabilities for network communication with Google Play APIs and local file system access for reading and writing listing data.
Sanitization: While the skill includes a validate step, this is restricted to structural linting (e.g., character limits and image dimensions) and does not provide sanitization for malicious prompt instructions contained within the listing content.

gplay-metadata-sync