pytorch-lightning
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: LOW
Full Analysis
- NO_CODE (SAFE): No malicious code or patterns were detected in the provided templates.
- PROMPT_INJECTION (SAFE): The instructions are standard documentation for deep learning workflows and do not contain any attempt to bypass AI safety guidelines.
- DATA_EXFILTRATION (SAFE): No sensitive file access or unauthorized network operations were identified.
- EXTERNAL_DOWNLOADS (SAFE): No remote code execution or untrusted downloads are present.
- REMOTE_CODE_EXECUTION (SAFE): The skill contains standard Python code for deep learning with no use of unsafe
eval(),exec(), or untrusted subprocess calls.
Audit Metadata