onnx-inference

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements standard machine learning workflows using well-known, industry-standard libraries such as ONNX Runtime, PyTorch, and Scikit-learn.
  • [SAFE]: Model management code, including the model registry and optimization logic, utilizes local file operations for JSON metadata and SHA-256 integrity hashing, following secure development practices.
  • [SAFE]: Cross-platform implementations for Rust and Go use established community wrappers (e.g., the 'ort' crate and 'onnxruntime_go' module) relevant to the skill's stated purpose.
  • [SAFE]: No obfuscation, prompt injection, or unauthorized network operations were detected; the skill's use of tools like Bash and Write is consistent with its role in software deployment and model export.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 02:51 AM
Security Audit — agent-trust-hub — onnx-inference