roblox-change-control
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill serves as a safety configuration guide, explicitly defining boundaries to prevent the AI agent from performing high-impact actions. It establishes strict rules against pushing code to production branches, running live synchronization tools that could incur costs, and deleting dormant code that might be linked to live products.\n- [COMMAND_EXECUTION]: Provides a set of diagnostic shell commands using grep, sed, jq, ls, and git. These are intended for the agent to audit the development environment and verify deployment triggers. All commands are read-only and targeted at internal project files, presenting no risk of unauthorized modification or exfiltration.
Audit Metadata