ai-ui-heuristics
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely composed of markdown documentation and reference materials. There are no scripts, binaries, or automated tools included, making it structurally safe from code-based attacks.
- [PROMPT_INJECTION]: The static analysis warning for instruction override in
references/nielsen-10-ai.mdis a false positive. The document provides legitimate, task-specific instructions for UX auditing and does not attempt to bypass agent safety filters or redefine core system behaviors. - [EXTERNAL_DOWNLOADS]: The skill contains a reference link to a well-known industry research group (Nielsen Norman Group). There are no remote code downloads, automated fetching of scripts, or suspicious external dependencies identified.
Audit Metadata