ai-sdk

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs running curl against https://ai-gateway.vercel.sh/v1/models at runtime to fetch current model IDs and then choose the model (which directly controls which model/instructions the agent uses), so this external URL is a runtime dependency that controls agent behavior.