level-design
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [Prompt Injection] (SAFE): No instructions attempting to bypass safety filters or override core agent behavior were detected. The content uses standard persona framing for a level designer.
- [Data Exposure & Exfiltration] (SAFE): No credentials, sensitive file paths, or network commands (like curl or wget) are present.
- [Unverifiable Dependencies] (SAFE): The skill does not install any third-party packages or download remote scripts.
- [Indirect Prompt Injection] (LOW): The skill relies on external reference files (patterns.md, sharp_edges.md, validations.md) to guide its logic. This introduces a surface where malicious instructions in those external files could influence the agent, although no such instructions exist in this file.
- Ingestion points: references/patterns.md, references/sharp_edges.md, references/validations.md
- Boundary markers: None provided to distinguish reference data from instructions.
- Capability inventory: No command execution or network capabilities found.
- Sanitization: None present.
- [Dynamic Execution] (SAFE): No runtime code generation or dynamic loading techniques are used.
Audit Metadata