ui-ux-tester
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The instructions describe routine software quality assurance activities, focusing on functional flows, visual alignment, and logic validation. No malicious patterns such as credential harvesting, command execution, or persistence mechanisms were found.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to parse external documentation to enumerate testing flows, which represents a surface for indirect prompt injection. 1. Ingestion points: Input documentation and screen state definitions (SKILL.md). 2. Boundary markers: None present. 3. Capability inventory: The skill is limited by a read-only sandbox mode and lacks tools for network exfiltration or system modification. 4. Sanitization: No explicit input sanitization is defined. The risk is assessed as safe due to the restricted environment.
Audit Metadata