foundation-models
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- General (SAFE): The skill provides developer documentation and Swift code snippets for utilizing Apple's native AI frameworks. No malicious instructions or hidden payloads were found.
- Data Privacy (SAFE): The framework focuses on on-device processing ('No cloud, no API keys'), which minimizes the risk of sensitive data exfiltration.
- External References (SAFE): All external links point to official Apple Developer documentation at developer.apple.com, which is considered a trusted domain.
- Indirect Prompt Injection (SAFE): While the framework facilitates LLM interactions, the skill itself is purely instructional and does not introduce an exploitable surface in the agent's runtime environment.
Audit Metadata