web-design-guidelines-vercel

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the skill to "Fetch the latest guidelines" from the public URL https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md using WebFetch before each review, meaning the agent will read and act on live, third-party GitHub-hosted content that can change its rules and outputs.

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 from https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md and uses the fetched content to control review rules and the output-format instructions, so the external content directly controls agent behavior and is a required dependency.