Creating & Configuring Agents

Creating a New Relational Agent

From Scratch vs. Templates

When you start creating a new Relational Agent, you can choose one of the pre-filled templates or start from scratch.

Templates

A good choice for newcomers. Templates come pre-configured and ready to use. Each template has its own Template Variables — a set of fields you fill in to tailor it to your context.

From Scratch

If you don’t find a template that fits your use case, you can build an agent from scratch.

Agent Evaluation

Optionally attach an evaluation to the agent. Use the search field to find and select an existing evaluation, or create one in Evaluations.

Agentic Purpose

What is Agentic Purpose for?

This is where you tell the agent what it’s supposed to do, how it should behave, and what path to follow in a conversation. Think of it as writing a job description and a script — combined.

It has three parts:

  • Objective — the goal. What is this agent trying to achieve? What should the user walk away with?

  • Guardrails — the rules. What topics or behaviours are off-limits? What should the agent always or never do?

  • Business Process Steps — the flow. The ideal step-by-step path through the conversation, from opening to close.

The agent uses all three together. It will follow the steps, stay within the guardrails, and keep the objective in mind throughout — adapting its tone and approach based on how the conversation unfolds.

All text fields here support rich text formatting, headings, lists, bold, and more. Well-structured instructions are easier for the agent to follow.

Rich text formatting toolbar

Objective

Define the primary and subsidiary objectives.

  • Primary objectives are broad and overarching

  • Sub-objectives break down how the agent achieves the primary goal

Example:

•

Your primary objective is to inspire the client to love our brand.

•

Your job consists in making the clients familiar with your brand’s intentions and the most recent product offer.

•

In order to do your job well, you should inquire into the client’s needs and present the brand in a fresh way.

•

Avoid pushing sales too hard if the client is hesitant. Focus on personalized engagement.

Guardrails

Rules and boundaries agents must obey.

  • Cover sensitive topics requiring expertise (e.g., psychological treatment, financial advisory)

  • Ensure non-standard customer requests are safely redirected back to the business process path

When two or more languages are selected in Allowed Languages, do not add a guardrail that restricts the agent to speak only one language. This conflicts with the multilingual setup and can cause problems with speech recognition.

Business Process Steps

This is the "golden path" of the ideal interaction. Write it step-by-step.

Step 1

Greet the client and explain the advantages of your offer.

Step 2

Learn more about the client’s needs.

Step 3

If the client seems interested, propose the best-fit product. If not, inspire interest with unexpected product features.

Step 4

Direct the client to more information and take a memorable farewell.

Assets Referencing

You can reference project assets by name using @mentions in Objective, Guardrails, and Business Process Steps.

Knowledge Bases

Ground the agent’s responses in specific content.

Interactive Content

Trigger structured interactions like forms, images, or web views.

Integrations

Call external services during or after the conversation.

Skills

Define what is being trained. In the Objective, reference Skills to establish the skill catalog for the session. In the Business Process Steps, reference a skill to select what is being practised and route to the right scenario. Only Skills available in your project can be referenced.

  • Global assets – Available automatically. Mentioning them explicitly is optional but improves performance.

Step 5

At the end of the interaction, ask the user how they were satisfied and display the multimodal @product_feedback_form

Relational Settings

Determine how strictly the Relational Agent follows the business process vs. focusing on relationship-building.

Task Focus

Strict execution of process. Example: Mortgage procedure with no deviation allowed.

Relational Focus

Conversation-driven, lighter process enforcement. Example: Brand loyalty and empathy-driven conversations.

Balance

Combines task and relational approaches. Example: Informing clients about new offers while still executing specific sub-goals.

Remembers users (memory control)

What it is

A simple toggle that controls whether a Relational Agent should recall previous conversations with the same user.

When to use it

Disable for one-off, shared, or public interfaces (e.g., kiosks) where each conversation should be treated as new. Enable for relationship-based use cases (e.g., account management, support follow-up) where continuity matters.

How to configure

Go to Agent configuration → Agentic Purpose → Remembers users.

  • On (enabled) – The agent remembers and can refer to past interactions

  • Off (disabled) – Every conversation is treated as a new interaction

Agent-specific setting. For concepts, policies, and governance, see see Memory & Relational Intelligence.

Agent’s Identity

Avatar

Select the look of your Relational Agent from the available avatar presets.

Environment

Select the environment in which the Relational Agent appears. The environment sets the visual background for the interaction.

Camera Mode

Control how the camera behaves during a conversation.

Dynamic

The camera moves during the interaction for a more cinematic feel.

Static

The camera stays fixed, similar to a video call.

Voice

Select the voice you want your agent to speak with. Each voice has its own set of supported languages, which determines the options available in Allowed Languages.

Allowed Languages

By default, all languages supported by the selected voice are enabled. If you don’t need all of them, remove the ones you don’t want.

Voice and Allowed Languages

Your voice supports American English, Czech, and German, but you only want the agent to speak American English and Czech. Remove German from Allowed Languages.

When a conversation starts, the platform matches the agent’s allowed languages against the Language setting on the user’s device (app) to decide which language the agent speaks:

  • The primary language is the highest-priority match between the two lists.

  • Any remaining matches become additional languages the agent can also speak.

The agent’s allowed languages are American English and Czech, and the user’s app is set to Czech, English:

  • Primary language → Czech

  • Additional language → American English

Character & Backstory

Describe the personal traits of your digital employee.

Elements can include:

  • Character traits, style of expression, and backstory

  • Previous experiences, likes/dislikes, behaviors in specific situations

  • Personal views on topics

A well-written character and backstory aligned with the agent’s role makes interactions feel natural and memorable.

Revisions

Every time you save changes to your agent’s configuration, a new revision is automatically created. This allows you to track changes and revert to previous versions if needed.

For details on managing revision states and reverting, see Testing & Publishing.