The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the evaluate_matlab_code tool to run the MATLAB buildtool utility. This allows the agent to execute a sequence of tasks (such as code analysis, testing, and packaging) defined within the project's buildfile.m. This functionality is central to the skill's purpose as an automated build executor.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it executes code defined in a local project file without an intermediate human review checkpoint. * Ingestion points: The skill loads and parses the buildfile.m located in the project root. * Boundary markers: There are no explicit markers or instructions used to prevent the agent from following instructions potentially embedded within the build file content. * Capability inventory: The skill is configured to use the evaluate_matlab_code tool, which provides the capability to execute the logic within the loaded build plan. * Sanitization: No sanitization or validation of the build file's contents is performed before the execution phase.

matlab-build-toolbox