The Agent Skills Directory

[SAFE]: The skill defines a structured methodology for game design, including greenfield development and existing-game audits. No malicious behaviors, exfiltration attempts, or obfuscation were identified across any of the analyzed files.
[INDIRECT_PROMPT_INJECTION]: The skill includes an audit mode that ingests external project data, which is an inherent part of its functional design for game analysis. There is no evidence of this being used for malicious purposes.
Ingestion points: references/application-playbook.md (§ 2. Existing-game audit) specifies inspecting actual project code, catalogs, scene documents, and tests.
Boundary markers: No specific markers or delimiters are defined for the data being analyzed.
Capability inventory: The skill uses jgengine-* tools for system implementation and workflow for shipping changes.
Sanitization: No explicit sanitization or filtering of external code content is mentioned.

game-design