unity-development
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- Prompt Injection (SAFE): The persona definition and core principles align with standard instructional patterns and do not attempt to bypass agent safety filters or override system constraints.
- Indirect Prompt Injection (SAFE): The skill utilizes internal reference files (references/patterns.md, references/sharp_edges.md, references/validations.md) to ground its responses. 1. Ingestion points: Internal references/ directory. 2. Boundary markers: The agent is instructed to treat these files as the 'source of truth'. 3. Capability inventory: None; the skill contains no executable code or network capabilities. 4. Sanitization: Not applicable as the files are internal to the skill package.
- Data Exposure & Exfiltration (SAFE): No access to sensitive data, hardcoded credentials, or network exfiltration patterns were identified in the skill content.
Audit Metadata