scenario-roblox-animation
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Python's subprocess module in rbx_animation.py to interface with local development tools. It executes Blender in headless mode (blender -b) to process animations and the Lune runtime to perform offline auditing of Luau scripts. These calls use specific, pre-defined command-line arguments and target bundled scripts.
- [INDIRECT_PROMPT_INJECTION]: The skill processes metadata from the Roblox Studio environment, such as rig hierarchies, joint configurations, and script source code. This data is ingested by the AnimRig and AnimAudit modules to provide technical analysis to the agent. The ingestion surface is managed via structured data and regex-based filtering of source code (e.g., stripping comments).
- [EXTERNAL_DOWNLOADS]: The documentation references the author's official GitHub repository for manual installation of sibling skills, which is a well-known and expected practice for this vendor's ecosystem.
Audit Metadata