search
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external metadata from app stores, which represents an indirect prompt injection surface.
- Ingestion points: External data from App Store and Google Play is ingested via aso_search_results and aso_search_suggestions in SKILL.md.
- Capability inventory: The skill can modify the user's AppFollow account using add_app, remove_app, and add_collection as seen in SKILL.md.
- Boundary markers: No specific delimiters are defined to separate untrusted search results from the system prompt.
- Sanitization: The skill trims the search query but does not sanitize the metadata returned from the store APIs.
Audit Metadata