tick
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: A thorough review of all skill files and references found no malicious patterns. The skill provides technical templates and documentation for game engineering best practices.
- [PROMPT_INJECTION]: The skill is designed to ingest and process data from other specialized agents (such as game design documents from 'Quest' or state machines from 'Weave'). While this represents a surface for indirect prompt injection, it is managed as a core functional requirement of the agent's role in a multi-agent system and does not currently leverage dangerous capabilities.
- Ingestion points: The skill reads external design artifacts, balance models, and prototypes (SKILL.md).
- Boundary markers: There are no explicit delimiters or 'ignore' instructions for external data, but the skill strictly limits its scope to implementing runtime systems based on those artifacts.
- Capability inventory: The skill is limited to code generation and architectural implementation; it does not contain subprocess calls, network exfiltration, or file-write operations beyond standard agent output.
- Sanitization: External content is not sanitized before being used to generate game systems.
Audit Metadata