Borlabs Cookie keeps its configuration in dedicated tables: wp_borlabs_cookie_groups for consent groups (Essential, Statistics, Marketing, External Media, Functional), wp_borlabs_cookie_cookies for each cookie record per group, wp_borlabs_cookie_services for the service definitions, and wp_borlabs_cookie_content_blockers for the embed blockers (YouTube, Vimeo, Google Maps, custom). Per-visitor consent is stored in the borlabs-cookie cookie as a structured JSON value.

SleekAI maps the full stack, so the system message can include the active groups, the cookies they own, the blockers in play, and the visitor's current consent state. "What's blocked until I accept External Media?" returns the list from wp_borlabs_cookie_content_blockers. "What does the Statistics group include?" returns the actual cookie list with retention. "Have I accepted Marketing?" reads the visitor cookie and answers honestly.

Display conditions scope the bot per template, multibot keeps consent help separate from product support, and conversation logs capture every privacy question with model, token usage, and origin page. SleekAI reads Borlabs Cookie; Borlabs Cookie still owns every consent write, content-blocker unlock, and group toggle.

Borlabs Cookie is one of the strictest consent stacks in WordPress, popular across DACH and EU sites that take GDPR seriously. It does more than show a banner. It blocks YouTube embeds, Vimeo, Maps, and custom embeds behind per-group consent, with optional per-embed acceptance for the cautious.

That power creates a UX problem: visitors see placeholders, get confused, ask why a video they expected to watch is showing a card. A generic chatbot has no idea the placeholder is a Borlabs content blocker, no idea which group it belongs to, and no idea whether the visitor already accepted Statistics. So it deflects, and engagement drops.

SleekAI fixes the gap by reading the same data Borlabs already stores: groups, services, cookies, blockers, and the visitor's borlabs-cookie JSON. With that context in the prompt, the bot can explain why a specific embed is blocked, which group accepting will unlock it, what other embeds that group would also enable, and how the per-embed accept differs. Compliance teams gain a feedback loop, because conversation logs show which groups confuse visitors most and which placeholder copy needs work.

None of this requires the bot to touch consent state. SleekAI reads, Borlabs writes. Every accept, decline, per-embed unlock, and group change stays with the plugin where the audit trail is canonical.

Combined with multibot, a privacy-help bot can live on heavy-embed templates while a separate product bot stays focused on the rest of the site. Add display conditions for admin-only DPIA review and the same approach scales from public consent help to internal compliance work, without leaking restricted data.