No. SleekAI is configured to refuse counselling, mediation, communication exercises, and any attempt to take sides between partners. The boundary holds even when the visitor explicitly asks the bot to help them work through the argument, because that work has to come from a licensed counsellor who can hear both partners in session. The system prompt enforces the refusal consistently across every conversation regardless of phrasing, and the bot redirects calmly while offering to help book a consultation instead.

Your system prompt should surface the National Domestic Violence Hotline at 1-800-799-SAFE (7233) before any other flow when signs of domestic violence appear. The bot is configured to recognise indicators like fear of a partner, controlling behaviour, isolation, or descriptions of physical or emotional abuse, and to lead with the hotline and 911 for immediate danger. Booking and practice information are secondary to safety, and the system prompt enforces that order explicitly. The bot does not improvise on this point.

Distress signals route to 988 (the US Suicide and Crisis Lifeline) before any other flow. The system prompt is configured to recognise signs like hopelessness, self-harm ideation, or descriptions of crisis, and to lead with 988 and the local emergency line before continuing with booking or practice information. Safety comes first, practice details second. Most practices also add language directing the visitor to nearby emergency services in case the distress is acute and 988 is not the right entry point.

Yes, when each counsellor is tagged in custom fields. The bot reads training lineage into the prompt and matches conversationally. Visitors often arrive with a preference (typically because their referring therapist or a friend uses specific language), and the bot routes accordingly. Visitors with no preference get a brief framing of the differences from the practice's published copy, with the decision about approach left to the counsellor in session, where it belongs.

Yes. Discernment Counseling is a scoped intervention for couples where one partner is uncertain about staying in the marriage, and it has a different intake pattern from standard couples work. The bot can route to Discernment-trained counsellors specifically and explain that the work is short-term (typically 1 to 5 sessions) and focused on the leaning-in or leaning-out decision rather than reconciliation. The system prompt handles the framing without overpromising what the intervention is for.

Yes. SleekAI is multibot, so each counsellor's bio page can run a chat scoped to their modality, populations served, and availability. This matters in couples work because Gottman, EFT, and Imago practitioners have different intake language, and a visitor reading the EFT counsellor's bio benefits from a chat that does not try to also represent the Gottman approach in the same prompt. Each bot has its own tone, intake script, and routing path.

Conversations are stored in your WordPress database. OpenAI or Anthropic calls use your own API key, so there is no SleekAI middleman server. Retention is set through the plugin, typically 7 to 30 days for couples and mental health practices given the sensitivity of conversation content. Most practices also forward conversation summaries to their EMR through a webhook and purge raw logs on a strict schedule that fits the privileged nature of the inquiries.

Yes. The 28+ display conditions let you scope the chatbot to booking-intent pages (contact, fees, counsellor bios, new couples) and hide it on the blog and longer educational posts. Visitors reading articles about communication patterns or attachment styles are often researching rather than booking, and forcing chat onto those pages can feel intrusive. For couples practices specifically, most firms also hide the bot on the about page where the visitor is comparing practices rather than scheduling.