The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (search.py, core.py, design_system.py) to search through its design database. A thorough analysis of these scripts shows they only utilize standard library modules (csv, re, pathlib, etc.) to perform BM25 text ranking on local CSV files. No unauthorized system commands, shell spawning, or network operations were found within the executable logic.
[INDIRECT_PROMPT_INJECTION]: The skill processes user queries to provide the agent with design recommendations. While this involves processing user input to retrieve data that influences the agent's output, the data source is entirely local and static. The risk is minimized as no external, untrusted content is fetched or executed.
[EXTERNAL_DOWNLOADS]: The skill documentation provides manual instructions for users to install Python (e.g., using brew, apt, or winget) if it is missing from their system. These are standard setup procedures for developer tools and do not involve automated or suspicious background downloads.

ui-ux-pro-max