frontend-design
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to execute external packages including
modern-web-guidance,axe-core, andpa11yusingnpx. These are well-known tools from reputable sources, including the Google Chrome team, for performing browser compatibility and accessibility audits. - [INDIRECT_PROMPT_INJECTION]: The
review-uiworkflow ingests and processes external UI code files (HTML, CSS, TSX). This creates a vulnerability surface where malicious instructions could be embedded in the data processed by the agent. - Ingestion points: The agent reads user-provided files during the UI review process as defined in
SKILL.md. - Boundary markers: The skill does not explicitly define delimiters or warnings to ignore instructions embedded within the code files being reviewed.
- Capability inventory: The skill uses
npxfor audits and performs file reads, but it does not have the capability to write files or execute the ingested UI code itself. - Sanitization: There is no evidence of sanitization or filtering of the external code content before it is processed by the agent.
Audit Metadata