gamedev-roblox
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill incorporates a mechanism for persistent learning by calling a local script. Evidence: SKILL.md instructs the agent to run a local script for feedback. Context: This is a legitimate workflow for maintaining state.
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for setting up a development environment using official tools. Evidence: references/setup-and-toolchain.md guides the user to install tools such as Rojo and Rokit. Context: These are standard industry tools.
- [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for external content through technical sources. Ingestion points: data/sources.json URLs. Boundary markers: Absent. Capability inventory: Local toolchain execution. Sanitization: Not implemented. Note: Sources are limited to highly trusted domains.
Audit Metadata