Skills
Navigate
skills/alfredang/skills/mobile-ios-design/Gen Agent Trust Hub

mobile-ios-design

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill includes a reference implementation for an iOS feedback form that ingests untrusted user input and processes it through a system-level action.
  • Ingestion points: references/ios-navigation.md defines a FeedbackView that uses TextField and TextEditor to capture arbitrary user-supplied title and message strings.
  • Boundary markers: Absent; the skill does not suggest specific boundary markers or delimiters to protect against prompt injection within the user-provided text.
  • Capability inventory: The implementation in references/ios-navigation.md includes the UIApplication.shared.open capability, which is used to transmit the captured data to an external service (WhatsApp).
  • Sanitization: The code follows best practices for secure URL construction by using URLComponents and URLQueryItem to ensure all user input is properly percent-encoded before being embedded in the target URL.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 07:30 AM
Security Audit — agent-trust-hub — mobile-ios-design