web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches design guidelines from Vercel Labs' official GitHub repository at raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md. This resource provides the rules and formatting instructions for the agent's audit.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and follows instructions from an external URL and local user-provided code.
  • Ingestion points: SKILL.md specifies fetching instructions from a GitHub URL and reading files matching user patterns.
  • Boundary markers: None present. There are no delimiters or instructions telling the agent to ignore prompt injection attempts within the fetched guidelines or user files.
  • Capability inventory: The skill utilizes file reading and network fetching tools.
  • Sanitization: No sanitization or validation of the fetched content is performed before the agent treats it as executable instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 08:45 PM