The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-generated content from Google Play Store reviews, which is a known vector for indirect prompt injection. Ingestion points: The skill fetches reviews via gplay reviews list, view, and history (SKILL.md). Boundary markers: The documentation explicitly instructs the agent to 'quote or summarize' content and treat reviews as data. Capability inventory: The agent can perform public writes using gplay reviews reply. Sanitization: The skill recommends using --dry-run for manual operator review of drafted replies.
[COMMAND_EXECUTION]: The skill executes the gplay CLI tool to perform its operations.

gplay-reviews