web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions found that attempt to override agent behavior or bypass safety filters.
  • [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, credentials, or network exfiltration patterns detected.
  • [Unverifiable Dependencies & Remote Code Execution] (SAFE): No external packages or remote scripts are downloaded or executed. The referenced repository (vercel-labs/agent-skills) is a trusted source.
  • [Dynamic Execution] (SAFE): No runtime code generation or execution identified.
  • [Metadata] (SAFE): Metadata is descriptive and consistent with the content of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:59 PM