Audit History
transformers - 6 audits
Audit version 6
Latest SafeJan 21, 2026, 07:41 PM
This skill contains only documentation and reference materials for the Hugging Face Transformers library. All static findings are false positives: backtick detections are markdown code blocks in documentation, eval references are PyTorch model.eval() method calls, and network references are documentation URLs. No executable code, command execution, or security risks present.
Risk Factors
⚙️ External commands (391)
⚡ Contains scripts (3)
🌐 Network access (3)
Audit version 5
Medium RiskJan 17, 2026, 03:15 AM
AI analysis failed after multiple attempts - MANUAL REVIEW REQUIRED before publishing. This skill cannot be auto-published until reviewed by a human.
Risk Factors
⚙️ External commands (391)
⚡ Contains scripts (1)
🌐 Network access (3)
Detected Patterns
Audit version 4
Medium RiskJan 17, 2026, 03:15 AM
AI analysis failed after multiple attempts - MANUAL REVIEW REQUIRED before publishing. This skill cannot be auto-published until reviewed by a human.
Risk Factors
⚙️ External commands (391)
⚡ Contains scripts (1)
🌐 Network access (3)
Detected Patterns
Audit version 3
SafeJan 7, 2026, 01:07 AM
Pure documentation skill containing only Markdown reference guides for Hugging Face Transformers library. No executable code, scripts, or potentially harmful capabilities. All capabilities shown (pip install, model loading, training) are standard ML workflows using trusted sources.
Audit version 2
SafeJan 7, 2026, 01:07 AM
Pure documentation skill containing only Markdown reference guides for Hugging Face Transformers library. No executable code, scripts, or potentially harmful capabilities. All capabilities shown (pip install, model loading, training) are standard ML workflows using trusted sources.
Audit version 1
SafeJan 7, 2026, 01:07 AM
Pure documentation skill containing only Markdown reference guides for Hugging Face Transformers library. No executable code, scripts, or potentially harmful capabilities. All capabilities shown (pip install, model loading, training) are standard ML workflows using trusted sources.