The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted external data from app listing pages, creating a potential surface for indirect prompt injection.
Ingestion points: Listing metadata including titles, descriptions, and reviews are fetched from apps.apple.com and play.google.com using the WebFetch tool as described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore commands potentially embedded in the scraped store content.
Capability inventory: The agent's actions are restricted to data scoring and report generation; it does not have access to sensitive file systems, persistent storage, or arbitrary command execution.
Sanitization: There is no evidence of filtering or sanitizing the retrieved listing text before analysis.
[EXTERNAL_DOWNLOADS]: The skill fetches data from well-known mobile application platforms.
The skill uses automated tools to retrieve live metadata and visual asset information from official Apple App Store and Google Play Store domains for analysis.

aso