ux-critique
Pass
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown-based instructions, checklists, and legal reference material intended to guide an agent in performing manual UX audits.
- [NO_CODE]: Analysis confirms the absence of any scripts, binaries, or automated execution instructions. The skill relies solely on the agent's reasoning capabilities based on the provided text guides.
- [PROMPT_INJECTION]: The instructions include rigorous controls like a mandatory filtering pass and a specific 'evidence contract' to prevent the agent from being influenced by deceptive interface patterns or manufacturing findings. No malicious instruction overrides were detected.
- [DATA_EXFILTRATION]: There are no instructions for accessing sensitive files or performing network requests, precluding any risk of data exfiltration.
Audit Metadata