API Catalog

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

• Potentially malicious external URL detected (high risk: 1.00). The skill calls OpenAI at runtime (e.g., POST https://api.openai.com/v1/chat/completions) to parse user voice commands and then uses the model's response to construct and execute Home Assistant actions, so external content from that URL directly controls agent behavior.