The Agent Skills Directory

[PRIVILEGE_ESCALATION]: The skill instructs the agent to use sudo for installing system utilities (apt install meshlab admesh), which requires administrative privileges and grants the process elevated permissions.
[COMMAND_EXECUTION]: The skill relies on executing several shell commands to process and verify files, including admesh for mesh repair and grep/head for verifying generated G-code metadata.
[EXTERNAL_DOWNLOADS]: The skill installs the meshlab and admesh packages from the operating system's standard package repository.
[INDIRECT_PROMPT_INJECTION]: The skill processes external 3D model files, which creates an attack surface for indirect prompt injection.
Ingestion points: The skill reads external CAD and mesh files (STEP, STL, OBJ, 3MF) provided via the source_model input path.
Boundary markers: There are no specific instructions or delimiters used to prevent the agent from interpreting content within the model files as instructions.
Capability inventory: The skill uses Bash to run repair and verification tools and Write to export sliced G-code files.
Sanitization: The input files are not sanitized or validated before being passed to system processing tools.

prepare-print-model