web-design-guidelines

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

• Third-party content exposure detected (high risk: 0.95). The required workflow explicitly fetches “fresh guidelines” from a public URL at runtime (outsider-authored web content) and then uses that fetched text as the rules/output-format instructions, which would be ingested into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs fetching runtime rules from https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md which are then used to control prompts and output format, so the external content directly controls the agent at runtime.