The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were identified during the analysis of the skill's instructions, scripts, and data files.
[COMMAND_EXECUTION]: The Python scripts (core.py, design_system.py, search.py) are used for local data processing and file management. They do not invoke shell commands or use any dangerous functions such as os.system() or subprocess.run().
[DATA_EXFILTRATION]: The skill does not perform network operations. It reads guideline data from local CSV files and persists output to a local design-system/ directory. No exfiltration of sensitive user data or system credentials was observed.
[PROMPT_INJECTION]: The instructions in SKILL.md guide the agent through a structured design workflow. There are no attempts to bypass safety filters or override the agent's core behavioral constraints.

ui-ux-pro-max