The Agent Skills Directory

[SAFE]: The skill is entirely focused on visual design quality and does not exhibit any malicious patterns or security risks. All operations are aligned with its stated purpose of UI polishing.\n- [COMMAND_EXECUTION]: The skill utilizes the grep utility to search for visual design tokens such as font sizes, spacing values, and hex colors in frontend source files. These search patterns are hardcoded, read-only, and do not involve shell injection or dangerous command parameters.\n- [PROMPT_INJECTION]: No patterns associated with prompt injection or behavior overrides were detected. The instructions maintain professional boundaries and guide the agent to perform specific design audits without attempting to bypass safety protocols.\n- [DATA_EXFILTRATION]: The skill does not perform any network operations or external data transfers. Its scope is limited to local file system reads of frontend source code and generating a markdown report within the agent session.

ui-polish-review