No. SleekAI is configured to refuse therapy, counselling, or coping advice. It books consultations and answers practice questions only. The boundary holds even when the visitor explicitly asks the bot for communication tools or relationship advice, because that work has to come from a licensed therapist who knows the situation. The system prompt enforces the refusal consistently across every conversation regardless of phrasing.

Yes. Configure your system prompt to surface 988, the Crisis Text Line, or your local equivalent when distress is detected. The bot prioritises safety first, then practice information, then scheduling - in that order. Most practices add explicit language to the prompt that catches phrases like 'want to hurt myself' or 'don't want to be here' and routes to crisis resources before continuing any intake flow.

Yes, if you tag therapists with their modalities and populations in custom fields. The chatbot reads those into the prompt. Couples work commonly requires specific training (Gottman, EFT, Imago) that not every therapist on a group-practice team holds, and the bot matches conversationally rather than booking the visitor with an individual-only therapist who would have to refer the couple out anyway.

Conversations are stored in your WordPress database. OpenAI or Anthropic calls use your own API key. SleekAI doesn't operate a middleman server. You control retention through the plugin, typically 7 to 30 days for mental health practices given the sensitivity of the conversation content. Most practices also forward conversation summaries to their EMR through a webhook so the privileged detail does not persist in the chat table beyond the retention window.

Yes. The 28+ display conditions cover post type, URL, page template, user role, device, and time of day. For therapy practices specifically, most firms hide the bot on the blog (because reading therapy articles often is itself the help-seeking activity) and show it only on contact, fees, and therapist bio pages where the booking intent is real. The conditions are precise enough to scope the bot exactly where it adds value.

Yes. SleekAI supports multibot configurations - one bot per bio page or per clinical specialty. This is particularly useful in group practices where the partners practise materially different approaches: an EFT-trained couples therapist's bot does not need to discuss IFS, and a Gottman-trained therapist's bot needs different intake language than the EMDR trauma specialist's bot. Each bot can have its own tone, intake script, and routing path.

Yes, when you have a sliding-scale page. Sliding-scale framing is sensitive because the practice does not want to gatekeep but also does not want to attract clients who would otherwise pay full fee. The bot can quote your sliding-scale eligibility framework (e.g. household income brackets) without making the visitor feel scrutinised, then route to a clinician who has sliding-scale availability if that is what the visitor needs. Pricing transparency improves both fit and retention.

Yes. Telehealth-only practices have different intake needs from hybrid or in-person practices: state of residence (because therapists are licensed by state), platform comfort, and tech setup. The bot can branch on the new-client page to capture state first, then check whether your therapists are licensed there before continuing the intake. This avoids booking a consult that has to be cancelled when the licensing mismatch is caught manually later.