laws-of-ux-lou
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions explicitly warn the agent to treat artifacts like screenshots or live pages as untrusted and to ignore any instructions embedded within the designs being reviewed, which is a key defense against indirect prompt injection.
- [COMMAND_EXECUTION]: The skill uses a local script,
scripts/validate-output.py, to verify that the generated critique follows the prescribed format. Technical analysis of this script confirms it is benign, performing only string processing and regex validation using standard Python libraries without network access or sensitive file operations. - [SAFE]: All reference materials are standard markdown files providing context on UX principles, and the platform configuration files follow expected patterns for agent deployment.
Audit Metadata