Botmaker 3.0 is an agentic version of the platform. It is an AI that manages and resolves complete business processes, capable of deciding, taking actions, and executing specific tasks in external systems. Unlike previous versions, it is not programmed with rigid scripts or decision trees: you describe the goal in natural language and the agent navigates on its own through the steps needed to resolve it.

You can:

• Create autonomous AI agents that manage complete business processes.
• Coordinate several agents with an orchestrator that decides which one responds based on context.
• Add tools to your agents such as knowledge bases, MCPs, code actions, and integrations.
• Design automations that run on their own by time or by event.
• Incorporate human validation (Human in the Loop) at the critical points of the process.
• Work collaboratively across several users on the same agents and orchestrators.

Key concepts

Before moving on to the answers, it helps to be clear on these definitions, which you will find throughout the article and within the platform.

Agent: an AI-powered bot that understands a specific business goal and decides what to do step by step to resolve it. One agent equals one goal. If you need to resolve several goals, you use several agents coordinated with one another.

Orchestrator: the director of the team of agents in a channel. It receives the customer's message, consults the relevant agents, and picks the most suitable one to respond.

Channel: the messaging medium through which the customer arrives. For example: WhatsApp, Webchat, Instagram, Telegram, Slack, or Google Business Messages.

Workflow: each logical step the agent goes through to resolve its goal. For example: “requesting data,” “confirming an appointment,” or “charging.”

Resource: a tool the agent can use to resolve a workflow. There are several types: logics, knowledge bases, integrations (MCP), code actions, and external apps.

Logic: an ordered sequence of steps. It is a type of resource, useful when the sequential order matters.

Automation: a trigger that runs on its own by time or by event, without the customer messaging the bot. It lives outside the agent.

Human in the Loop (HITL): a way of working in which the agent knows how to ask a human operator to validate, confirm, or resolve an exception.

Multimodal: the agent understands and responds with text, audio, or image within the same conversation.

MCP (Model Context Protocol): a standardized connector to an external system such as a CRM, a calendar, or a database.

Ticket: a record the agent generates for each workflow started within a conversation. It lets you see the status of the process and follow up on it.

Costs and billing

How is the cost of using Botmaker 3.0 calculated?

Billing is based on token consumption. You pay for what your agent consumes: the more it resolves, the more it bills. If it is not used, you do not pay. The configuration tool has no extra costs: creating agents, orchestrators, automations, or entities does not generate a charge by itself. The cost appears when the agent runs and starts handling requests.

Is billing based on tokens, conversations, or messages?

On tokens, and the cost depends on the model. You will find the consumption detail on the platform's Account screen. The MAU or DAU session modalities continue to work the same way. Only the generative consumption is added.

Do orchestrators generate additional costs?

The orchestrator does not have an extra cost as a piece, nor is it billed separately. It is included in the plans. Its execution does consume tokens, just like any agent.

Will I be able to see the cost per agent?

Yes. Each feature reports its own tokens. You will be able to know how much each agent, each orchestrator, and each feature consumes, and attribute the spend by business area. The generative technologies screen shows the detail of tokens used by each agent, orchestrator, and by Human in the Loop.

AI models

Which AI model does Botmaker 3.0 use?

In this first version, the platform works with the default model Open AI 5.4. The model detail and its cost per consumption are available on the Account screen.

Will I be able to choose the model (LLM) I use?

We are working so that you can select the model at the level of each agent, so that a single orchestrator can coordinate agents with different models depending on the case (by token consumption or by reasoning capacity).

Agents, orchestrator, and workflows

What is an agent and what does it do?

An agent resolves a specific business process. Think of it as a virtual employee with a well-defined job. For example, in a dental center you could have one agent to book appointments, another to answer frequently asked questions, another to handle dental emergencies, and another to send satisfaction surveys. Each agent has a goal, its own workflow, and a set of tools to resolve it.

What is the orchestrator?

It is the new piece in version 3.0. It works as a layer that receives the customer's message, looks at which agents are available in that channel, and picks the most suitable one to respond. It asks the relevant agents (not all of them, to save tokens), evaluates which answer best fits the context, synthesizes a single response, and sends it.

How do channels, orchestrators, and agents relate to one another?

The relationships are simple:

• 1 channel = 1 orchestrator. Each channel has a single orchestrator. There cannot be two directors in the same conversation.
• 1 orchestrator = N agents. An orchestrator manages a team of several specialized agents.
• 1 agent = N orchestrators. A well-configured agent can be reused across several channels and orchestrators without needing to duplicate it.

What is a workflow?

It is the minimal unit of agent behavior. Each agent goes through a sequence of workflows depending on the conversation. Workflows can branch: the agent chooses the next one based on context, not based on a fixed tree. For example, the “Book appointment” agent could have the workflows: greeting + intent, request basic data, show availability, confirm appointment, and closing.

Can AI agents be combined with traditional scripted flows?

Yes. The paradigm has been inverted: you now work with orchestrated AI agents and, when you need finer control or a rule-based flow, you create it as a logic within the agent. The logic works as an ordered sequence of steps that the agent follows to resolve a specific task.

How do you give the bot a personality? Where is the tone configured?

The tone, style, restrictions, and use of buttons are configured at the orchestrator level, not in each agent. Each agent inherits that personality when responding, which prevents you from ending up with contradictory instructions between agents. In the orchestrator you can define:

• Tone: formal, informal, warm, or technical.
• Response length: short or detailed.
• Personality: brand voice (friendly, neutral, fun).
• Restrictions: what it cannot say or promise.
• Whether or not it allows buttons in the response.

How are guardrails configured, and how do you keep the agent from going off track?

Restrictions are defined at the orchestrator level. In addition, internally Botmaker works as a set of agents that check one another, so it is rare for an agent to stray from the territory it was created for. Agents only respond using the tools they have available: if a tool for a topic is not added to them, they do not engage in conversation about that topic.

Agent resources and tools

What tools (resources) can an agent use?

There are five types of resources available:

• Logics: an ordered sequence of steps. For when the sequential order matters. For example, the step by step of booking an appointment.
• Knowledge bases: PDFs, websites, FAQs. The agent reads and responds with that information. For example, accepted medical rates and coverages.
• Integrations (MCP): turnkey connectors to external systems such as Google Sheets, Google Calendar, CRMs, and databases.
• Code actions: code integrations the agent can run. Useful for custom integrations.
• External apps: other tools integrated into the platform.

How does the agent decide when to use each resource?

The decision is made by the model based on the field “When will it be used by the agent.” If the resource description is clear and specific, the agent knows when to invoke it. If it is ambiguous, it may get confused or call it when it shouldn't.

Note: an ambiguous description like “For appointment topics” usually generates errors. A clear description like “When the patient requests a new appointment and has already provided their data” works much better.

How do logics work?

Logics are built with three types of blocks:

• Instruction: a generic indication to the agent (“Ask the patient for their ID”). It is the most used.
• Conditional: branches based on conditions (“If the coverage is Swiss Medical, show premium schedules”). The most important in practice.
• Loop: repeat N times. Useful for processing lists; rarely used in conversation.

A good practice when building logics is to use several small boxes with one instruction each, instead of one giant box with several instructions chained together. The model processes separate instructions better.

Can I upload documentation or files as a knowledge base?

Yes. Knowledge bases accept PDFs, websites, and FAQ content. The agent reads that content and responds based on it.

Can code actions still be used?

Yes. Code actions remain available as one more resource the agent can turn to when it needs it. The code actions you already had configured in previous versions can still be used in Botmaker 3.0.

What are automations?

They are processes that run on their own, without the customer messaging the bot. There are two types:

• By time (schedule): they run at defined moments (every hour, every Monday at 9 a.m., the 1st of each month).
• By state (event): they react to something that happened in a conversation, such as a workflow change or a completion.

Some examples: sending an appointment reminder 24 hours before, loading data into the CRM when an appointment is confirmed, or sending a satisfaction survey a day after the consultation.

Available channels and integrations

Which channels does Botmaker 3.0 support?

Botmaker 3.0 supports the main messaging channels integrated into the platform. Among them are WhatsApp, Webchat, Telegram, Instagram, Facebook, and Google Business Messages.

Is it useful for automating internal processes, not just conversations with customers?

Yes. Botmaker 3.0 stands out as a conversational process platform, but it can also be used for non-conversational processes. You can design agents and automations that connect your ERP, your CRM, spreadsheets, and other tools. For example, invoice uploads, internal notifications, or administrative routines.

Multimodality

Do agents understand images, audio, or video?

Yes. The models are multimodal: they can understand text, images, audio, and video. Not only what an image says, but also its essence. For example, a customer can send a photo of a promotion via WhatsApp and the agent can understand the content to take the order.

Human in the Loop, tickets, and traceability

What is Human in the Loop (HITL) and when is it used?

Human in the Loop adds a way of working in which the AI agent stays in the middle of the conversation at the service of the human operator. The human asks the agent to confirm, execute, or develop a response, and the agent communicates with the end user. At the same time, the agent presents the human with the customer's communications along with analysis, suggestions, and shortcut buttons so that they only have to validate.

Any workflow state can be marked as “requires human confirmation.” When a case reaches that point, the AI stops and hands control to an operator, who sees the case in the dedicated Human in the Loop inbox. According to internal tests, this mode allows manual work to be reduced by up to 82%.

How do you see the conversations each agent is handling?

Every time an agent starts a workflow, a ticket associated with that conversation is generated. Tickets appear in a new section below Chats and are also accessed directly from the agent. There you see the workflow status, the ticket details, internal comments, the associated conversations, and the activity. You can also route the ticket to a support team or add followers so they stay informed.

Is there a Kanban-style view to see the tickets?

Yes. Tickets can be viewed as a list or as a Kanban-style board where the columns are the workflow states and the tickets are distributed according to which stage they are in. You can move a ticket from one state manually, as long as the conditions defined in the workflow are respected.

Can I see previous versions of an agent and revert changes?

Yes. Agents have a versioning system: every adjustment you publish is recorded in a version that you can view and revert if a recent change did not give the expected result.

Can you work collaboratively across several users?

Yes. The platform allows several people to work simultaneously on the same agents. Each user is identified with a color, you see who is online and where they are working, and you can divide tasks or work in parallel on different agents and orchestrators.

Remember to visit our Help Center for further information.