The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze 3D models based on user-provided descriptions and files, which represents an ingestion point for untrusted data.
Ingestion points: Phase 1b (Model Analysis) in SKILL.md instructs the agent to run an 11-point checklist on model descriptions or files provided by the user.
Boundary markers: While a mandatory reporting format is defined, there are no explicit delimiters or instructions for the agent to ignore embedded commands within the user data.
Capability inventory: The skill enables the generation of custom G-code (Phase 5), printer networking setup (Phase 8), and shell command execution for validation (Phase 10).
Sanitization: No input sanitization or validation logic is defined for the model descriptions.
[COMMAND_EXECUTION]: The skill provides specific shell commands for the agent or user to execute external validation tools that are not included in the provided skill files.
Evidence: SKILL.md (Phase 10) and references/profiles.md contain shell code blocks for running OrcaSlicer_profile_validator and orca_extra_profile_check.py. Since these scripts/executables are not bundled with the skill, they represent unverifiable external dependencies.

orca-slicer