ux
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the command
git diff --name-only HEAD~1in SKILL.md to identify UI files for review. This is a standard development workflow for context-aware auditing and does not present a security risk.\n- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface as it ingests untrusted code for review. However, it includes strong mitigation via Behavioral Rule 9, which mandates the detection and refusal of deceptive design patterns.\n - Ingestion points: Target UI files identified in Phase 1 (SKILL.md).\n
- Boundary markers: Absent.\n
- Capability inventory: Subprocess call to git (SKILL.md) and tool invocations for recording and charting.\n
- Sanitization: Implements an ethical override (Behavioral Rule 9 in SKILL.md) that explicitly rejects and flags dark patterns within the processed code.
Audit Metadata