The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill has a defined attack surface for indirect prompt injection by ingesting untrusted data from the Google Play Store.
Ingestion points: The skill uses gpd reviews list to retrieve user-provided review text and gpd vitals errors reports search to fetch crash logs and stack traces.
Boundary markers: The instructions do not specify any delimiters (e.g., XML tags or triple quotes) or provide guidance for the agent to treat external content as untrusted data rather than instructions.
Capability inventory: The skill provides the agent with high-impact capabilities, including publishing new app versions (gpd publish), issuing financial refunds (gpd purchases subscriptions refund), and modifying developer account permissions (gpd permissions grants create).
Sanitization: There is no evidence of sanitization, filtering, or validation of the text retrieved from reviews or vitals reports before it is processed by the agent.

gpd-cli