design-brand-guardian
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were detected. The skill contains standard documentation and utility scripts for design tasks.
- [COMMAND_EXECUTION]: The skill documentation includes examples for running local Python scripts (
check_contrast.pyandgenerate_tokens.py). These scripts use only Python standard libraries (argparse,json,colorsys,sys) and do not perform any dangerous operations like network requests, unauthorized file access, or system modifications. - [DATA_EXPOSURE]: The
check_contrast.pyscript includes functionality to read from a local JSON file provided by the user via the--fileargument. This is used solely for batch processing color pairs and does not expose sensitive system information. - [PROMPT_INJECTION]: The instructions in
SKILL.mdare focused on brand strategy and consistency. No patterns of prompt injection or attempts to bypass AI safety guardrails were found.
Audit Metadata