lov-professional-portrait

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and referenced documents do not contain any malicious patterns. The functionality is limited to generative image editing using the environment's built-in tools.
  • [DATA_EXFILTRATION]: The skill includes instructions to protect user privacy by explicitly telling the agent not to publish, upload, or add user portraits to a repository without explicit user consent. No unauthorized network transmissions or hardcoded credentials were detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided images. The instructions mitigate potential injection risks by mandating strict 'identity-preserve' prompt templates and quality gates that verify identity, facial geometry, and skin texture remain consistent with the original source, preventing the agent from deviating from the intended task.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 08:49 AM