frontend-design-review
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data, specifically design assets, specifications, and mockups. This establishes an attack surface where hidden instructions within these files could attempt to influence the agent's behavior.
- Ingestion points: SKILL.md (Workflow Phase 1: Gather Design Assets).
- Boundary markers: The skill does not provide delimiters or instructions to treat external design text as untrusted data separate from its primary guidelines.
- Capability inventory: The skill is entirely instructional and does not include scripts with file system, network, or shell capabilities.
- Sanitization: No mechanisms for filtering or sanitizing content from the ingested design files are described.
- [NO_CODE]: The skill consists entirely of Markdown documentation and reference guidelines. There are no executable scripts (Python, JavaScript, or Shell) or remote code execution patterns, which significantly reduces the potential for privilege escalation, persistence, or data exfiltration.
Audit Metadata