tinyworld-island-viewer
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions define internal logic for a game's sequential generation system. It covers entity placement (houses, paths, water, etc.), coordinate systems (8x8 grid), and state management (loading guards, rendering defaults).
- [COMMAND_EXECUTION]: The skill mentions standard development commands such as
npm test,npm run build, andnpm run stats:island-viewer. These are documented for validation and testing within a local project context and do not involve remote code execution or suspicious parameters. - [DATA_EXPOSURE_AND_EXFILTRATION]: All file paths referenced (e.g.,
scripts/island-viewer.js,styles/island-viewer.css) are relative to the skill's project structure. There are no hardcoded secrets, sensitive system paths, or network operations targeting external domains. - [PROMPT_INJECTION]: The 'Rules' section uses instructional language to define the behavior of a game engine, not to override the agent's safety protocols or system instructions.
Audit Metadata