The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted content from external App Store and Google Play listing pages. This metadata, which includes app descriptions and user reviews, represents an attack surface where malicious instructions could be embedded to manipulate the audit's findings or the agent's behavior.
Ingestion points: Listing data retrieved via WebFetch and visual data from listing page screenshots as instructed in SKILL.md.
Boundary markers: The prompt instructions do not specify the use of delimiters or 'ignore embedded instructions' markers when processing the retrieved listing content.
Capability inventory: The skill utilizes WebFetch for network retrieval of web pages and browser tools for capturing full-page screenshots.
Sanitization: There is no explicit requirement in the instructions to sanitize, escape, or validate the content of the app listings before it is interpolated into the agent's context for scoring.

aso