simulink-configure-real-time-model
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides instructions for the legitimate task of configuring Simulink models for real-time execution. It adheres to documented workflows and incorporates user-consent gates for high-risk actions like connecting to targets and saving files.
- [COMMAND_EXECUTION]: The skill utilizes the
evaluate_matlab_codetool to execute MATLAB-specific commands such asset_param,get_param, andload_system. These operations are standard for model management within the MATLAB environment. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from Simulink model files, which represents a potential attack surface. 1. Ingestion points: Simulink model configuration parameters and structure (SKILL.md). 2. Boundary markers: Explicit user approval is required before saving changes or establishing hardware connections (SKILL.md). 3. Capability inventory: The skill can modify model settings and update diagrams using standard model-edit tools (manifest.yaml). 4. Sanitization: Not applicable, as the skill interacts with structured model parameters rather than processing natural language from the model.
Audit Metadata