cryptokit

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEMETADATA_POISONING
Full Analysis
  • [METADATA_POISONING]: The skill provides multiple external links in the 'References' section and within the text that claim to be official Apple documentation but point to an unofficial domain (sosumi.ai). This is deceptive and could lead an agent or developer to rely on potentially modified or untrusted documentation for security-critical cryptographic implementations.
  • [METADATA_POISONING]: The skill includes instructions and availability checks for non-existent software versions (e.g., 'iOS 26+'). While this may be forward-looking, it provides inaccurate context for current development environments and could lead to the generation of code with incorrect or unreachable logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 11:13 AM
Security Audit — agent-trust-hub — cryptokit