unity-llm-integration
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- SAFE (SAFE): No security vulnerabilities were detected in the provided skill file. The instructions are focused on guiding the agent's behavior as a specialized Unity developer.
- PROMPT_INJECTION (SAFE): No evidence of jailbreak attempts or safety bypasses. The grounding instructions requiring the agent to use local reference files are standard for expert persona skills and do not attempt to override system-level safety protocols.
- DATA_EXPOSURE (SAFE): No sensitive file access, hardcoded credentials, or network exfiltration patterns were identified.
- EXTERNAL_DOWNLOADS (SAFE): No package installations (npm/pip), remote script downloads, or external network requests were found.
- INDIRECT_PROMPT_INJECTION (LOW): The skill mentions reading data from references/*.md. This ingestion surface is low risk as the files are internal to the skill package and the agent's capabilities are limited to providing text-based advice and code samples to the user.
Audit Metadata