The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content retrieved from App Store and Google Play listing pages via WebFetch. This data, which includes app descriptions and user reviews, creates a surface for indirect prompt injection attacks.
Ingestion points: Listing metadata and user reviews are fetched from apps.apple.com and play.google.com as described in SKILL.md.
Boundary markers: The skill contains an explicit instruction to ignore commands embedded in the fetched data.
Capability inventory: The skill uses WebFetch and screenshot tools to gather information.
Sanitization: The skill relies on natural language instructions rather than programmatic sanitization to handle untrusted input.

aso