web-design-guidelines

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs fetching and using live rules from an open GitHub raw URL (https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md) via WebFetch as part of its required workflow, so untrusted third-party content can directly control checks and output behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly fetches runtime guidelines and output-formatting instructions from https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md via WebFetch before each review, and that fetched content directly controls the agent's prompts and is required for operation.