vercel-queues

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow reads queue message payloads via handleCallback for push-mode consumers (and similarly via receive in poll-mode), and those message contents can originate from outsider-submitted publishes into the queue topics.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's CRITICAL note explicitly instructs fetching external docs at runtime ("Before writing queue code, fetch the docs at https://vercel.com/docs/queues"), so remote content from that URL would directly control the agent's instructions for code generation.