ui-design-token-refactor
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is focused on improving CSS architecture and design systems. All documentation and provided code align with this intended primary purpose.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to run a bundled Python script (
scripts/audit_style_tokens.py) to inventory style debt in a project. \n - The script uses only Python standard library modules (
argparse,re,pathlib, etc.). \n - It performs read-only regex matching on source files to identify hardcoded colors, dimensions, and Tailwind patterns. \n
- It implements proper directory filtering to avoid sensitive or irrelevant paths like
.gitandnode_modules. - [REMOTE_CODE_EXECUTION]: No remote code execution patterns or external script downloads were found.
- [DATA_EXPOSURE]: The skill does not access sensitive credentials, configuration files, or perform network exfiltration. File access is limited to auditing source code for styling tokens.
- [PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety filters, or extract system prompts were detected in the instructions or reference materials.
Audit Metadata