building-3d-objects

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: No security issues were detected. The skill follows established Roblox development patterns and focuses on physical object construction.
  • [COMMAND_EXECUTION]: The skill utilizes the mcp__roblox__run_code tool to execute Lua scripts within Roblox Studio. This capability is used for 3D modeling tasks such as part instantiation, CSG operations, and object validation. The provided validation script (scripts/validate.luau) contains benign checks on model properties such as anchoring, positioning, and materials.
  • [PROMPT_INJECTION]: The skill processes user-provided object descriptions to generate code, which constitutes an indirect prompt injection surface. However, the instructions are strictly constrained to geometric primitives and physical properties. Ingestion points: User building requests triggered by specific keywords in SKILL.md. Boundary markers: None explicitly defined in the Lua templates. Capability inventory: Lua code execution within the Roblox Studio environment via mcp__roblox__run_code. Sanitization: No input sanitization or validation of the user request content is performed before generating the build logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 12:30 PM