passkit

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes functions that ingest external binary data for the purpose of creating and updating Apple Wallet passes. This data processing creates a surface for indirect prompt injection if the source content is maliciously crafted.
  • Ingestion points: The addPassToWallet(data:) function in SKILL.md and the updatePass(newPassData:) function in references/wallet-passes.md are the primary entry points for untrusted data.
  • Boundary markers: No explicit delimiters or specialized prompt instructions are used to isolate the binary data input.
  • Capability inventory: The skill includes high-impact capabilities such as handling payment tokens via PKPaymentAuthorizationController and modifying the user's pass library via PKPassLibrary.
  • Sanitization: The skill relies on the standard PKPass(data:) initialization, which performs cryptographic signature verification and manifest validation, providing significant protection against unauthorized pass modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 02:16 AM