web-design-guidelines

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md explicitly requires fetching and ingesting rules from a public third-party URL (see "Guidelines Source" pointing to https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md and the "How It Works" steps), so untrusted, user-authored content is read and used to drive review decisions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly fetches and uses https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md at runtime to obtain the rules and output-format instructions that directly control the agent's review prompts and is required for operation.