No. SleekAI's system prompt is configured to share general planning concepts and your firm's services without recommending specific securities, projecting personalised retirement income, or making individualised tax or insurance recommendations. When a prospect asks something advice-shaped, the bot describes what would be covered in a planning engagement and defers the actual modelling to a CFP. The line is drawn in the prompt, reinforced by the guideline filter, and visible on every conversation log.

Yes. SleekAI reads your ADV Part 2A, Part 2B, and Form CRS as pages or PDFs on your WordPress site and passes the relevant language into the model context. The bot describes the fiduciary standard using your own disclosure wording rather than a generic definition, so what visitors see in the chat matches what they would see if they downloaded the document directly from the SEC's IAPD site.

Yes, at a general-education level. The bot explains that fiduciary means recommendations must be in the client's best interest at all times, contrasts that with the suitability standard that can apply to commissioned brokerage activity at fee-based hybrids, and notes that some advisors switch standards depending on which hat they are wearing. It will not characterise specific named firms or individual advisors, only the standards themselves.

Conversation logs sit in your WordPress database with model name, token usage, and page URL captured per session. Most fiduciary firms export them to their CRM or to an archival system like Smarsh or Global Relay on a recurring schedule via webhook so chat lives in the same archive as email. You control retention, who in admin can view them, and whether summaries get emailed to the advisor after each session.

Yes. SleekAI supports multibot deployment with 28 plus display conditions: post type, taxonomy term, user role, logged-in state, URL pattern. A retirement-focused bot can run on retirement pages, a different bot on business-owner planning, and a separate bot on the blog. Each bot has its own system prompt, knowledge sources, and disclosures so the framing matches the topic.

Claude Sonnet and GPT-4o both follow the no-specific-recommendation prompt reliably and produce consistent disclosure language. Smaller or older models tend to drift into advice-shaped statements when a prospect presses them, which is the exact failure mode a fiduciary cannot afford. SleekAI is bring-your-own-key for OpenAI, Anthropic, Google, or OpenRouter, so you choose the model that matches your tolerance.

Yes. The system prompt can encode any ideal-client profile (pre-retirees with concentrated equity, business owners contemplating sale, dual-income households above a threshold), and the bot uses that profile during the conversation to gauge fit. Prospects who clearly do not match get pointed at self-serve content or a referral, while qualified prospects get fast-tracked into the discovery flow with the conversation summary attached.

It explains what those products are at a general-education level and describes whether your firm includes them in planning engagements. It does not recommend specific products or carriers, that is a fit question requiring a credentialed conversation. If a prospect wants product-specific advice, the bot routes to an advisor. If your firm does not offer a particular product, the bot says so directly rather than improvising, which keeps the conversation aligned with what the firm actually does.