web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches design guidelines from Vercel Labs' official GitHub repository (vercel-labs/web-interface-guidelines) to serve as the rule set for its audits.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it incorporates instructions from an external file and processes user-provided code.
  • Ingestion points: Reads local UI code files and fetches the command.md file from a remote repository.
  • Boundary markers: Absent. The instructions do not specify delimiters to separate the retrieved guidelines or user code from the agent's core logic.
  • Capability inventory: The skill uses file-read capabilities and WebFetch for network operations.
  • Sanitization: No explicit sanitization or filtering of the content from the external guidelines or user files is performed before the agent interprets them.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 09:43 AM