Most yes. Text, textarea, number, select, checkbox, image, file, URL, date, and taxonomy fields all flow into context as text. Cloneable groups expand into structured lists. Map fields and color fields are passed as their stored values. Binary or proprietary field types are skipped silently, which is rarely an issue for chat use cases since those values are usually rendered into something readable by the time they hit the page.

Yes. If you have offloaded data to a custom table through the MB Custom Table extension, register the table in the SleekAI Wizard and pick which columns the bot should see. The Wizard handles the join back to the post and exposes the row as a named variable in the system prompt, so high-volume data like variants or inventory does not need duplication into postmeta.

Meta Box stores relationships in a dedicated mb_relationships table rather than postmeta. SleekAI reads that table through the same Wizard interface, so the bot can follow a product to its accessories or a course to its lessons. You can choose to inline the related items or expose them as a separate variable referenced by the system prompt.

No. New fields appear in the Wizard automatically as long as they are registered through Meta Box. Map the ones you want the bot to see, save the chatbot, and the next conversation includes them. There is no embedding step and no separate training cycle, so iteration on field structure stays fast as the schema evolves.

Yes. Multibot mode runs several chatbots on one site with independent system messages, data sources, and display conditions. A common Meta Box setup is one bot for the product CPT and another for the documentation CPT, both coexisting under the same SleekAI install with their own field mappings and system prompts.

Yours. SleekAI is bring-your-own-key, so message costs are billed directly by your provider with no SleekWP markup. Use OpenAI, Anthropic, Google, or OpenRouter, and pick a fast model for catalog questions and a stronger one for complex reasoning under the same chatbot without changing the integration.

Yes. SleekAI reads the rendered page, so any value you display through a Meta Box View becomes context for the bot. For FrontEnd Submission, the bot can run on the form template and reference the in-progress submission's fields, useful for guided application or quoting flows where users want help while filling out the form.

Yes. Every conversation is logged inside WordPress with the page URL, user role, model name, token usage, and full transcript. The logs surface which fields draw the most questions, which often points to field labels or descriptions that should be clearer on the page itself, not just in the bot's system prompt.