This page covers how to configure a Lyra coach's name, description, purpose, personality, icebreaker message, and supported languages.
On this page: Agent name, purpose, and description | Purpose | Personality | Icebreaker message | Supported languages
Agent name, purpose, and description
These fields identify and display your coach to both admins and end users.
|
Field |
Purpose |
Visible to end users? |
|---|---|---|
|
Name |
The display name shown to users (e.g., "Sales Coach", "New Joiner Guide") |
Yes |
|
Description |
A short summary of what this coach does, shown to users when browsing available coaches |
Yes |
|
Personality |
Controls the tone and style of interactions |
No, it drives how the coach interacts with learners |
Tip: Choose a name and description that make it immediately clear what the coach is for. Users are more likely to engage with a coach called "Sales Discovery Coach" with the description "Practise consultative discovery with structured questioning frameworks" than one called "Coach 1".
Purpose
The Purpose field defines the core role and behaviour of the coach. This is the instruction that shapes how the coach responds — its tone, scope, and approach. Think of the Purpose as a job description for the coach. It tells Lyra: who are you, what do you do, and how should you behave?
Why it matters: A well-written purpose is the single most important factor in coach quality. Vague purposes produce vague responses. Specific, well-scoped purposes produce coaches that feel genuinely useful.
Writing a good purpose
-
Be specific about the coach's role and boundaries. Tell the coach exactly what it should and shouldn't do. A coach that tries to do everything will do nothing well.
-
Be user-centric. Frame the purpose around what the user needs, not what the coach is. "Help new joiners navigate their first 90 days" is better than "You are an onboarding AI".
-
Reference your knowledge base. If the coach should draw on specific types of content (e.g., company policies, sales playbooks), say so explicitly in the purpose.
-
Set the tone. Should the coach be formal or conversational? Directive or exploratory? Coaching-style (asking questions to guide reflection) or encyclopaedic (giving direct answers)?
Purpose examples
Personality
The Personality setting controls the tone and style of your coach's interactions. It shapes how users perceive the coach, whether it feels friendly and conversational, formal and authoritative, or something in between.
A clearly defined personality ensures a consistent experience that matches your organisation's voice and meets your audience's expectations. A leadership coach might be reflective and measured; a sales coach might be direct and energising.
Lyra offers four predefined personality options based on best practices. These provide a quick, reliable starting point. You can select a predefined personality under the Personality tab of your coach configuration. If the predefined options don't match your needs, select Custom personality to define your own.
Icebreaker message
The Icebreaker Message is the opening message users see when they first interact with a coach. It sets the tone for the conversation and tells the user what the coach can help with.
Why icebreakers matter more than people think: They are often the only thing that gets a first-time user to engage. A weak icebreaker ("Hi, how can I help?") leads to a vague first message, a poor first conversation, and a user who doesn't come back.
Write instructions, not scripts. Rather than writing a fixed welcome message, write an instruction that tells the coach how to greet users. This means the coach can personalise the greeting, draw examples from the knowledge base naturally, and stay consistent with its personality across sessions.
A good icebreaker instruction should:
-
Greet the user and establish the coach's role in one or two sentences
-
Give 3–4 short examples of things the user can ask about, drawn from the knowledge base
-
Frame the examples in the user's voice ("How do I…", "What should I…") rather than the coach's voice ("I can help with…")
-
End with an open question inviting the user to share what's on their mind
Icebreaker examples
Supported languages
Lyra coaches can respond in the following languages:
|
Code |
Language |
|---|---|
|
EN |
English |
|
ES |
Spanish |
|
FR |
French |
|
DE |
German |
|
IT |
Italian |
|
PT |
Portuguese |
The chat interface adapts automatically to the user's browser language preferences. If the browser language is not in the supported list, the interface defaults to English. The coach will attempt to respond in the same language the user writes in.