aso
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data fetched from external App Store and Google Play listings via WebFetch. An attacker could potentially embed malicious instructions within an app description to influence the agent's behavior during the audit. This is a common surface for indirect prompt injection in agents that consume web content.
- Ingestion points: Listing data retrieved from
apps.apple.comandplay.google.com(documented in Phase 1 ofSKILL.md). - Boundary markers: No specific delimiters are used to wrap the fetched listing content before analysis.
- Capability inventory: The skill utilizes WebFetch, screenshot tools, and local file read access (
.agents/product-marketing.md). - Sanitization: No explicit sanitization or filtering of the fetched text is performed before processing.
Audit Metadata