White-label AI agent that takes actions, not just answers questions

There is a ceiling on what you can charge for a chatbot that only answers questions. Clients see it as a fancy FAQ page, a convenience rather than a necessity, and they price it accordingly in their heads — maybe $100, maybe $150 a month. Useful, but not transformative. The moment the AI starts taking actions — looking up orders, booking appointments, processing returns, checking account balances — the perceived value shifts from "nice-to-have" to "this replaces a support agent." That shift is the difference between a $150/month chatbot service and a $400/month AI agent service. For agencies building a reselling business, the distinction between chatbot and agent is not semantic. It is the difference between modest margins and margins that justify building a team around.

Custom Tools are the mechanism that makes this possible. Available on Standard ($139/month) and Pro ($449/month) plans, Custom Tools let you configure the AI to call your client's own API endpoints during a live conversation. You specify the endpoint URL, the HTTP method, the parameters the AI needs to collect from the visitor, and a plain-language description of what the tool does. The AI handles everything else: it recognizes when a visitor's question or request matches a configured tool, asks for the required information naturally ("Could you share your order number so I can look that up for you?"), makes the API call, and presents the result in conversational language. The visitor never sees a loading screen, a redirect, or a form. They ask a question and get an answer — or ask for an action and see it completed — all within the same chat window.

The use cases span every industry your agency serves. For ecommerce clients, the AI agent checks order status, retrieves tracking numbers, initiates returns, and looks up product availability. A customer who would have submitted a support ticket or waited on hold gets their answer in fifteen seconds. For appointment-based businesses — dental practices, salons, consulting firms, fitness studios — the agent checks the scheduling system for open slots and confirms bookings inside the conversation. For SaaS clients, the agent looks up account details, checks subscription status, resets passwords, and triggers support workflows. Each of these actions currently requires a human to handle a support request, an email, or a phone call. The AI agent collapses that process into a single conversational exchange.

The architecture keeps everything isolated between clients, which is critical when you are managing Custom Tools that connect to live business systems. Each of your 20 sites on the Pro plan operates independently: separate knowledge base, separate Custom Tools configuration, separate conversation history, separate analytics. The ecommerce client's order lookup tool has no connection to the dental practice's appointment booking tool. API credentials, endpoint URLs, and parameter definitions are stored per site. When you onboard a new client, you configure their tools from scratch — no risk of inheriting another client's endpoints or exposing one client's data to another's visitors. The isolation is not just logical; it is structural. Twenty sealed environments under one account.

The pricing leverage for agencies is substantial. A basic chatbot service — one that answers questions using the client's website content — commands $150 to $250 per month. That is the market rate for a Q&A widget, and pushing above it is difficult because the client can always point to free or low-cost chatbot alternatives. An AI agent with Custom Tools is a different product entirely. When you demonstrate that the agent can check a customer's order status in real time, book an appointment without human intervention, or process a return request end-to-end — those are tasks that currently require staff time. A single support agent costs $3,000 to $5,000 per month in salary. An AI agent that handles 40% of those interactions for $300 to $500 per month is a straightforward cost-reduction pitch. You are no longer competing with free chatbots. You are competing with hiring.

Setting up Custom Tools for a client does not require you to build APIs or write backend code. The client's existing systems almost always have the necessary endpoints. Shopify stores have order APIs. Booking platforms like Calendly, Acuity, and SimplyBook have scheduling APIs. CRMs like HubSpot, Salesforce, and Zoho have contact and deal APIs. Payment processors have transaction lookup APIs. Your role as the agency is to identify which actions the client wants the AI to perform, locate the relevant API endpoints in their existing tools, and configure them as Custom Tools in the dashboard. The configuration is done through a form interface — endpoint URL, method, parameters, description — not through code. A typical client setup with three or four Custom Tools takes thirty to forty-five minutes, most of which is spent understanding the client's workflow rather than configuring the platform.

The knowledge base and Custom Tools work together, not in isolation. When a visitor asks "What is your return policy?", the AI searches the knowledge base (built from the client's website content) and delivers the policy. When that same visitor then says "OK, I want to return my order #7291," the AI switches from knowledge base lookup to Custom Tool execution — calling the returns API with the order number, checking eligibility, and initiating the process. The transition is seamless from the visitor's perspective. They asked a question, got an answer, made a decision, and the AI acted on it — all within a single conversation. This combination of knowledge and action is what makes the AI an agent rather than a chatbot. It understands context and it does things.

White labeling the AI agent means your clients never know about the upstream platform. On the Pro plan, the "Powered by Asyntai" badge is removed. Your agency name appears in the chat header. Your logo loads in the widget. Your brand colors define the visual identity. The client-facing dashboard — where your clients review conversation transcripts (including which Custom Tools the AI used and what results were returned), view analytics, and export leads — is embeddable on your website via a JS snippet and carries your logo and brand name. When a client logs in to check how many orders the AI agent looked up this week or how many appointments it booked, they see your dashboard, your brand, your product. The technology is invisible. The value is visible.

Language support extends to Custom Tool interactions, not just Q&A. The AI supports 36 languages with automatic detection. When a Spanish-speaking visitor asks about their order status, the AI collects the order number in Spanish, calls the tracking API, receives the response (typically in English from the backend), and presents the tracking information to the visitor in Spanish. The language layer wraps around the entire interaction — knowledge base answers, tool parameter collection, and tool result presentation. For agencies serving clients with international customer bases, this means a single AI agent handles visitors in any supported language without per-language configuration or separate tool setups.

The competitive moat for agencies offering AI agent services is deeper than for basic chatbot services. Once you configure Custom Tools for a client — connecting their order system, their booking platform, their CRM — and the AI agent starts handling real customer interactions, switching costs rise significantly. The client would need to reconfigure all of their tool integrations with a new provider, retrain the AI on their content, and accept downtime during the transition. More importantly, the AI agent accumulates interaction data that informs how well it handles edge cases: unusual order statuses, booking conflicts, return eligibility questions. This learned behavior compounds over time. A client who has been running an AI agent for six months has a more capable agent than one who just started. That capability gap is a retention mechanism that basic chatbots do not have.

The operational model for managing AI agent clients is lean because most of the ongoing work is handled by the platform. The knowledge base updates automatically when the client's website content changes. Custom Tools do not require maintenance unless the client changes their API endpoints. Conversation monitoring, analytics, and lead management happen in the dashboard. When a client wants to add a new Custom Tool — say, they launch a loyalty program and want the agent to check point balances — it is a fifteen-minute configuration task. You are not building integrations from scratch each time. You are adding a tool definition: endpoint, parameters, description. The AI figures out when and how to use it.

Agencies scaling beyond 20 sites or 50,000 messages per month move to the Custom Pro plan, which provides higher limits tailored to volume. The affiliate program offers a lighter-touch alternative: refer businesses to Asyntai with a tracking link and earn 20% recurring commission on their subscription without managing the setup, Custom Tools, or ongoing service. Some agencies blend both — white-label AI agents for their premium managed clients, affiliate referrals for businesses that want self-service. The two models serve different segments of your pipeline without conflicting.

The market timing is noteworthy. Businesses are actively looking for AI solutions that go beyond answering questions. They have seen basic chatbots and found them limited. They want AI that does things — checks orders, books appointments, resolves issues. Agencies that can deliver an AI agent under their own brand, configured with Custom Tools specific to each client's business, occupy a position that is difficult for competitors to replicate quickly. The combination of white-label branding, multi-tenant architecture, per-client Custom Tools, and 36-language support is not something a freelancer can stitch together from open-source components over a weekend. It is a productized service with real infrastructure behind it, available for $449 per month, ready to resell on day one.

Questions about Custom Tools, white-label AI agents, or reselling? Contact hello@asyntai.com or start with the Pro plan today.