game-generation-guidelines
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill establishes a rigid framework for AI-generated code, ensuring that resulting games adhere to a specific architectural pattern using approved libraries.
- [DATA_EXFILTRATION]: The guidelines include strict prohibitions against using
fetch()calls or external URLs for assets, which serves as a proactive defense against unauthorized data transmission or asset-based tracking. - [REMOTE_CODE_EXECUTION]: The skill limits the execution environment by restricting imports to a specific set of dependencies (
phaser,bitecs,colyseus, and internal@sdr/packages), preventing the introduction of unvetted third-party code. - [PROMPT_INJECTION]: While the skill processes external 'topic' data from a randomizer to influence game generation, it does not include instructions that would allow these inputs to override the primary safety or operational constraints of the agent.
Audit Metadata