webots-world-building
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides static VRML syntax examples for Webots simulation nodes such as Solid, Shape, and Physics. It contains no instructions that attempt to bypass AI safety guidelines or override agent behavior.
- [NO_CODE]: The skill explicitly forbids controller programming and API usage. It operates entirely within the scope of scene description, posing no risk of unauthorized command execution or system modification.
- [INDIRECT_PROMPT_INJECTION]: The skill documents the capability to import external assets via
MeshandCadShapenodes (e.g.,.obj,.dae,.stepfiles). - Ingestion points: External file references in
Mesh { url [...] }andCadShape { url [...] }nodes withinSKILL.mdandreferences/world_building_reference.md. - Boundary markers: None; the agent processes these as standard asset paths within the simulation context.
- Capability inventory: No execution capabilities detected; the skill only generates declarative scene tree text.
- Sanitization: Not applicable, as the skill produces static configuration rather than processing untrusted runtime input.
Audit Metadata