Qubely's 27 blocks (Row, Pricing, Accordion, Tabs, Testimonial, Team, Timeline, Post Grid, Contact Form, Pie Progress, Counter, and more) serialise into wp_posts.post_content as wp:qubely/* block markup that resolves to real HTML at render time. SleekAI reads the rendered page, so a Pricing block becomes plan rows with values, an Accordion becomes Q&A pairs, a Timeline becomes a list of events with dates, and a Post Grid becomes a small in-page index, all in the chatbot's context.

The data-source wizard pulls ACF fields, custom taxonomies, and any postmeta referenced by a block. So a Tabbed product page that mixes Qubely Tabs with custom spec fields can answer specific questions from both sources. Display conditions scope different bots by post type, taxonomy term, user role, or URL pattern, so a pricing-page bot, a team-page bot, and a docs-page bot can run side-by-side under multibot.

Qubely's row-column structure and shape dividers don't affect the bot since SleekAI reads post-render HTML. Conversations log inside WordPress with model name, token usage, and page URL. The JS API and PHP embed open the bot from a custom Qubely Button or Icon, with the current page already loaded as context, so a click on a Pricing block can open the bot with that block's data already in the prompt.

Qubely sites tend to be studios, agencies, and service businesses that pack a lot of trust-building content into a single page: pricing for several packages, team bios with experience, timelines, testimonials, and Q&A in an accordion. A generic chatbot ignores all of that. It guesses at package details, invents team members, and quotes timelines from training data that has nothing to do with the actual studio.

SleekAI reads the rendered Qubely blocks, so the bot quotes the Pricing block exactly, surfaces the right team lead from the Team block, and gives timeline answers from the Accordion FAQ. Because the read happens at request time, updating a package price or swapping a team member shows up in the next answer without retraining. Multibot mode lets a services bot, a team bot, and a contact bot share the install with separate prompts and scopes.

Conversation logging surfaces which packages or roles confuse visitors so you can either tighten the prompt or restructure the page. The end result is a chatbot that scales with the studio without becoming a maintenance burden.