agent-game-developer
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external input through the $ARGUMENTS placeholder. Ingestion points: SKILL.md (via $ARGUMENTS). Boundary markers: No specific delimiters or instructions are provided to separate the system prompt from user-supplied data. Capability inventory: Inherits capabilities of a general-purpose agent, which typically includes code generation and file system access. Sanitization: No evidence of sanitization, filtering, or validation for the incoming arguments.
- [NO_CODE]: The skill consists only of instructional text and metadata without any accompanying scripts or executable files, which significantly limits the potential for direct system compromise.
Audit Metadata