getitune-running-inference

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill involves loading and executing machine learning models in various formats (.xml, .onnx, and PyTorch checkpoints). Loading model weights, especially using framework-specific formats like PyTorch checkpoints, is an inherent part of the inference process but involves the dynamic loading of data structures.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to process external datasets and model files, which constitutes a potential attack surface.\n
  • Ingestion points: Dataset and model file paths provided in SKILL.md.\n
  • Boundary markers: None present.\n
  • Capability inventory: File system read access and execution of inference engines (engine.predict, engine.test).\n
  • Sanitization: None present.\n- [EXTERNAL_DOWNLOADS]: The documentation includes a reference to the official ModelAPI repository hosted on GitHub by the vendor (open-edge-platform). This is a legitimate resource reference for technical documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 01:43 PM
Security Audit — agent-trust-hub — getitune-running-inference