apple-appstore-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data (iOS source code and metadata) to provide feedback. This creates a surface for indirect prompt injection if the audited code contains malicious instructions.
- Ingestion points: Audits codebase files including
Info.plist, entitlements, privacy manifests, and StoreKit configurations. - Boundary markers: None identified in the prompt templates.
- Capability inventory: Generates reports, prioritized recommendations, and App Store Connect reviewer notes.
- Sanitization: No explicit sanitization of the codebase content is performed before processing.
- [EXTERNAL_DOWNLOADS]: The instructions reference the official Apple App Store Review Guidelines website. This is a well-known and trusted technical resource.
- [SAFE]: The skill instructions are focused on compliance and optimization, with explicit constraints preventing the agent from modifying code during its initial assessment.
Audit Metadata