web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on Mar 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches design guidelines and rule definitions from Vercel Labs' official GitHub repository to provide runtime instructions for UI auditing.
  • [DATA_EXFILTRATION]: Accesses local UI source code files to perform the requested audit; this file access is restricted to the scope of the review task.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted content from both external URLs and local files to drive its reasoning engine.
  • Ingestion points: Remote guidelines URL (SKILL.md) and user-specified UI code files.
  • Boundary markers: None explicitly defined to separate instructions from data.
  • Capability inventory: File system read access and network fetching (WebFetch).
  • Sanitization: No validation or sanitization is performed on the fetched guidelines or target code files before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 28, 2026, 05:37 PM