web-design-guidelines

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly instructs the agent to use WebFetch to fetch and apply live guidelines from the public URL https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md, and states the fetched content contains rules and output-format instructions that are used to drive checks and outputs, so untrusted third-party content can directly influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill fetches https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md at runtime and uses the fetched file as the authoritative rules and output-format instructions that are injected into and directly control the agent's prompts, making it a required remote instruction source.