scenario-3d-worlds
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows standard instructional patterns for interacting with AI models and tools. It provides a guide for using various 3D generation tiers (Marble, HY World, TripoSplat).
- [COMMAND_EXECUTION]: The document mentions a command for users to install sibling skills:
npx skills add scenario-labs/skills --skill <name>. This is an explicit instruction for the user to perform an action using the vendor's own official repository (scenario-labs), which is a legitimate and safe administrative action within this ecosystem. - [INDIRECT_PROMPT_INJECTION]: Like any skill that facilitates processing user-provided content (prompts, images, videos) through external models, there is a theoretical surface for indirect injection if those inputs contain instructions. However, the skill does not exhibit unsafe interpolation practices and focuses on providing parameters to established tool schemas, which is standard behavior for an agent skill.
Audit Metadata