web-design-guidelines

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs the runtime workflow to fetch guidelines from a public URL (outsider-authored web content) and then use that fetched text as the rules/context for the review, creating an indirect prompt-injection path via the fetched page content.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs the agent to fetch and use the remote rules file at https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md at runtime to control its checking rules and output format, so the fetched content directly controls agent behavior.