No. The system prompt is configured to describe packages, fee bands, and process while deferring every specific tax-treatment, entity-election, and deduction question to a credentialed CPA or EA. The bot will explain what your year-end package covers and quote the band. It will not opine on whether a particular meal is deductible or whether the owner should elect S-corp status. Those questions get routed to a partner with the conversation context attached.

SleekAI itself runs in WordPress and does not connect to QBO or Xero directly. What it does well is describe your QBO or Xero workflows from your service pages, ask the prospect which platform they are on, and hand off a scoped lead to your team with that detail captured. The actual books integration happens in your accounting workflow, not in the chatbot.

Yes. Add fees as a single page, a custom post type, or ACF fields, and SleekAI reads them through your existing WordPress structure and passes them into the prompt as context. The bot quotes the same numbers your engagement letters do, every conversation. Update the page and the next conversation reflects it without any redeploy or retraining step.

The bot asks how many months are behind, what software the books live in, and whether prior-year tax returns are filed. It then quotes the per-month catch-up rate (typically $150 to $250 per behind-month) and a target turnaround, and books the partner call. The catch-up scope ends up in the conversation log so the partner walks into the call already knowing the size of the job.

Bring your own key for OpenAI, Anthropic, Google Gemini, or OpenRouter, plus any OpenAI-compatible endpoint such as Azure or self-hosted models behind a compatible proxy. Most small-business firms run GPT-4o or Claude Sonnet because those models follow strict no-specific-advice system prompts more reliably than smaller models, which matters on a firm site where the cost of a hallucinated answer is reputational.

Yes. SleekAI ships with 28+ display conditions so the firm can show the bot only on service pages, hide it on the team blog, or trigger different bots per practice area. A common setup runs a bookkeeping-focused bot on the bookkeeping page, a tax-focused bot on the year-end page, and a generalist bot on the homepage that triages to the right specialist after the first qualifying question.

Yes. The system prompt can encode the patterns for each type of small business, and the bot adjusts its intake checklist based on what the prospect describes. A landscaper gets asked about contractor 1099s and equipment depreciation; a Shopify store gets asked about sales tax and inventory; a SaaS gets asked about R&D and Stripe payouts. Same firm, same bot, different qualifying questions per prospect.

Conversation logs live in your WordPress database, which means the firm controls retention, who can view them in admin, and whether transcripts get emailed to the partner after each session. Many small-business firms purge logs monthly and forward summaries to their practice management tool via webhook so client files stay clean and compliance reviews are easy to run.