web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • EXTERNAL_DOWNLOADS (LOW): The skill fetches instructions from https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md. As per the [TRUST-SCOPE-RULE], this is a trusted source (Vercel), which downgrades the severity of the remote fetch.
  • PROMPT_INJECTION (LOW): This skill is vulnerable to Indirect Prompt Injection (Category 8) because it retrieves instructions from an external URL to guide its logic.
  • Ingestion points: The command.md file from the remote Vercel repository.
  • Boundary markers: None specified; the skill instructions simply state to "Apply all rules from the fetched guidelines."
  • Capability inventory: The skill has the capability to read local files provided by the user.
  • Sanitization: No sanitization or validation of the fetched content is performed before applying it to the agent's context.
  • DATA_EXFILTRATION (SAFE): The skill performs a GET request to a trusted domain to retrieve data. There is no evidence of local data being transmitted to external endpoints.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:34 PM