The Agent Skills Directory

[SAFE]: The skill provides structured workflows for testing Simulink components using official products like Simulink Test and Requirements Toolbox. All external references are directed to trusted MathWorks domains for licensing and documentation.
[SAFE]: The skill's instructions incorporate security best practices, such as mandating the use of absolute paths for file operations to ensure artifacts are created and accessed only within the intended working directory.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for indirect prompt injection as it processes user-provided Gherkin (.feature) files and requirement specifications (.slreqx).
Ingestion points: Data enters the system via model_test (feature files) and the test_read/test_run family (test and requirement files).
Boundary markers: The Gherkin dialect uses explicit TOML front-matter blocks to separate configuration from test logic.
Capability inventory: Capabilities are limited to standard file system operations, model simulation, and test reporting within the MATLAB/Simulink environment.
Sanitization: The skill includes internal helper scripts (e.g., sanitizeDescription.p) designed to process and validate metadata and descriptions before they are used in test generation.
[DYNAMIC_EXECUTION]: The skill uses the evaluate_matlab_code tool to execute internal package functions (vnv.internal.agentic.*). This execution is limited to the skill's own distributed logic for tool orchestration and does not incorporate arbitrary external code.

testing-simulink-models