matlab-model-ams-systems

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses MATLAB to generate and execute simulation scripts at runtime to build Simulink models. This behavior is the primary intended function of the skill.
  • [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by extracting text from user-supplied PDF datasheets to define model parameters. * Ingestion points: User-provided PDF files processed via the extractFileText function (Phase 1.0 in SKILL.md). * Boundary markers: None implemented for the extracted text. * Capability inventory: The skill has capabilities to write files, execute generated MATLAB code, and run Simulink simulations. * Sanitization: Extracted text is used as technical data for parameter derivation without specific sanitization.
  • [DYNAMIC_EXECUTION]: The included diagnostic scripts probeSlxModel.m and quickProbe.m utilize evalin to inspect and extract variables from Simulink model workspaces for debugging and analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 12:12 PM
Security Audit — agent-trust-hub — matlab-model-ams-systems