Kid-safe, engaging chat for ages 4-8. The service uses Groq for LLM responses with guardrails that classify intent using the full conversation history, ask clarifying questions for unsafe/bad intent, and answer safe informational queries in a brief, child-friendly style.
