Urgent infant routing is the first rule in the system prompt. Lethargy, unresponsiveness, very pale colour, fewer wet diapers than expected for the baby's age, progressing jaundice, or weight loss beyond expected triggers an immediate referral to the family's pediatrician on-call line and to 911 in active emergencies. The bot does not assess infant safety in chat. This rule applies before any feeding mechanics discussion happens and is non-negotiable, even when the visitor was originally asking about latch.

No. Tongue tie evaluation is a clinical assessment done in person by an IBCLC, dentist, or ENT, and a frenectomy decision is a clinical and procedural decision. The bot describes what your tongue tie consult involves, who on staff conducts it, and what referral pathways look like to your partner dentists or ENTs. It never tells a parent their baby has tongue tie from chat. That answer is harmful and quickly wrong.

Yes. Publish each visit type as a page with what it includes, how long it lasts, the price, the radius for in-home, and the technical requirements for telehealth. The bot reads from those pages and recommends the appropriate visit type based on the baby's age and the concern. Newborns with latch concerns usually route to in-home, older babies with pumping or weaning questions route to telehealth, follow-ups can go either way.

Only as you publish it. List each consultant's credential (IBCLC, CLC, CLE, IBLCE-track) and what your practice has them doing. The bot reads scope from your published copy rather than from any built-in regulatory model, which keeps the answers honest and current. If your practice has both IBCLCs and CLCs with different visit types, the bot routes accordingly based on the published assignment.

Publish your insurance policy. Most IBCLCs are out of network and provide superbills, with some PPO plans reimbursing the full amount directly to the family. A growing number of insurers (Aetna and BCBS in some states) pay IBCLCs in network through The Lactation Network or similar. The bot quotes your policy from your published page and is explicit that coverage verification happens before or at the visit, never as a chat promise.

Yes, if it has a JS API. SleekAI's JS API lets the chatbot push intake details into your scheduling tool so the visit arrives prefilled. Practices on Acuity, Calendly, and Healthie all run this pattern. Smaller practices often have the bot create a WordPress custom post with the intake fields that the practice owner reviews and converts to a booking manually. Either pattern works.

Yes, if you set the booking widget to expose same-day availability. For newborns with latch concerns, same-day or next-day is often the right answer, and the bot can prioritise that based on the baby's age captured at intake. For older babies and follow-ups, standard booking is fine. The bot reads urgency rules from how you've configured your services, not from any built-in clinical assumption.

Conversations live in your WordPress database. Model calls go to OpenAI, Anthropic, Google, or OpenRouter using your own API key, never through any Sleek-hosted service. Retention is configurable. For an IBCLC practice that handles PHI, treat the chatbot as an administrative tool and avoid asking for detailed clinical history in chat. The intake should capture enough to route the visit and prepare the IBCLC, not enough to be a clinical record.