foundation-models-on-device
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely informational, providing Swift code templates for developers. It does not contain any executable files or external dependencies.- [PROMPT_INJECTION]: The provided code patterns include example instructions that incorporate safety guardrails for the model, such as instructions to refuse dangerous requests.- [DATA_EXFILTRATION]: All described operations occur on-device; no network-based data transmission or secret exposure was found in the provided patterns.
Audit Metadata