higgsfield-3d-cgi
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional, providing templates and technical vocabulary for 3D rendering. It does not execute arbitrary code, access sensitive system files, or perform unauthorized network operations.
- [SAFE]: Asset generation is gated by a mandatory user review and confirmation step as specified in the 'Prompt-building workflow' and 'Opt-in generation' sections. This prevents the agent from autonomously consuming credits or generating content without explicit consent.
- [SAFE]: The toolset (
mcp__higgsfield__*) is limited to the Higgsfield ecosystem for media generation and account management, which is consistent with the skill's stated purpose. - [SAFE]: Metadata and reference materials are professional, consistent, and do not contain any obfuscated content or malicious prompt injection patterns.
Audit Metadata