web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on Apr 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill retrieves design guidelines and formatting instructions from Vercel Labs' official GitHub repository (vercel-labs/web-interface-guidelines). This is a trusted source used to provide the latest design rules for the UI audit.\n- [PROMPT_INJECTION]: The skill analyzes untrusted user-provided files as its primary function. While processing external data is a prerequisite for its design review purpose, this creates a potential surface for indirect prompt injection. However, given the instructions are fetched from a trusted source and the data processing is transparently related to the skill's utility, the risk is minimal.\n
  • Ingestion points: User-specified files or patterns (SKILL.md)\n
  • Boundary markers: Not explicitly defined\n
  • Capability inventory: Reading local files and fetching external content via WebFetch (SKILL.md)\n
  • Sanitization: No specific sanitization of file content is defined before analysis
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 3, 2026, 04:37 AM