Yes. SleekAI reads your carriers directory directly, so the bot can name UPS, FedEx, Old Dominion, Estes, Saia, Averitt, Southeastern, and any other carrier you have in network. The directory can also be filtered by mode (LTL, TL, parcel, intermodal, ocean), region, and lane, and the system prompt can encode rules like 'when the buyer asks about LTL Southeast, list the top five carriers by transit performance'. Updates to the directory propagate automatically; no retraining or redeployment needed.

Yes, if you publish it. The bot reads your EDI documents page, which typically lists supported transaction sets (850, 856, 810, 940, 944, 214, 990, 997) and any partner-specific variants for Walmart, Target, Costco, Amazon, and similar large retailers. EDI is one of the highest-friction parts of logistics evaluation, and a bot that can confirm 'yes, we support Walmart's 856 variant including the SSCC-18 labeling spec' compresses the integration-feasibility conversation substantially.

Yes. The system prompt can include rules like '3PL operators route to the 3PL-platform team, shippers route to the shipper-platform team, brokerage shops route to the brokerage team'. The bot asks one qualifying question (operator type or shipment volume) and routes accordingly. Hand-off is via your existing CRM intake form, so the bot does not hold prospect data in any Sleek-hosted system, and the right SE owner gets the conversation transcript attached.

The bot reads your integrations directory, which describes the actual WMS and TMS integrations (Oracle Cloud WMS, Manhattan Active Warehouse Management, Blue Yonder, MercuryGate, BluJay, Manhattan Active Transportation, Oracle TMS, and so on). Each integration has a posture (native, API, EDI-bridge, partner-built), and the bot can quote the posture in one sentence. The system prompt should explicitly forbid claiming an integration that is not in the directory, because false integration claims surface in implementation as failed pilots.

Yes. SleekAI is multibot, so the shipper-facing surface can run a shipper-aware bot, the 3PL-facing surface can run a 3PL-aware bot, and the umbrella site runs the general bot. Display conditions scope each bot to the right URL pattern, post type, or UTM source. Shipper and 3PL evaluation criteria diverge enough (3PLs care about white-label, multi-tenant, and rating-engine flexibility; shippers care about coverage and rates) that a single umbrella bot ends up muddled, and multibot keeps the answers crisp.

The marketing-site bot does not generate live freight quotes, because rates depend on lane, equipment, mode, accessorials, and current capacity. The bot can describe the rating engine, the supported modes, and the typical turnaround for a quote, then route the prospect to the actual quote tool. Generating quotes from the chatbot would create commitment exposure for rates the system might not actually be able to fulfill, so the right pattern is to describe the capability and route to the quote tool.

Yours. SleekAI is BYO API key, so you bring your own OpenAI, Anthropic, Google, or OpenRouter key. For a typical logistics startup marketing site the monthly model cost is usually $30 to $200 in actual API usage depending on traffic. Logistics buyer content tends to be dense (carrier lists, EDI matrices, integration directories), so a stronger model is often worth the extra cost on enterprise-shipper or 3PL-pilot qualification, where the conversation can run long with technical specifics.

On your WordPress install. You control retention and exports, and you can pipe interesting conversations to a Slack channel for the GTM team via webhook. Logistics enterprise buyers often have strict data-handling expectations driven by their own customers (large retailers, regulated manufacturers), and the fact that prospect conversations never touch a third-party SaaS except the LLM API you chose is a simpler story to defend than a hosted chat tool. Logs include model name, token usage, and page URL for attribution.