apple-foundation-models

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFE
Full Analysis
  • [Overall Security] (SAFE): The skill is entirely informational, containing Markdown documentation and Swift code snippets for developer reference. No malicious behavior, obfuscation, or unauthorized data access was detected.
  • [Data Exposure] (SAFE): The documentation emphasizes local-first execution for privacy-sensitive data and provides guidance on when to escalate to cloud-based models versus keeping data on-device.
  • [Indirect Prompt Injection] (SAFE): While the described APIs (SystemLanguageModel) naturally handle user-provided text which could contain injections, the skill provides standard implementation guidance and mentions the existence of built-in system safety guardrails.
  • [External Downloads] (SAFE): All referenced URLs point to trusted official Apple developer and research domains. No unverifiable dependencies or third-party script downloads are present.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 03:28 AM