Settings

Settings is where you configure your Lyro workspace - who has access, how your AI agents behave, and how Lyro classifies, formats, and connects to the rest of your stack. Each area lives on its own tab.

Team

The Team tab is where you manage who can access your workspace.

• Invite a teammate by email - they receive an invitation and appear as pending until they accept.
• Resend or revoke a pending invitation at any time.
• Remove teammates, or leave a workspace yourself.

Every member holds one of two membership roles:

Tip: Invite the people who handle escalated conversations early so they are ready when Lyro hands work over. See Escalation.

Roles

The Roles tab defines reusable role instructions for your AI agents. A role is a named block of guidance - for example "Support specialist" or "Onboarding guide" - that you can attach to an agent to shape how it responds.

• Create a role with a name and a set of instructions.
• Edit or delete the roles you create.
• Built-in roles are read-only and cannot be changed or removed.

Pair roles with agent configuration in Agents.

API Keys

The API Keys tab manages credentials for programmatic access.

• Workspace API keys authenticate server-to-server calls to the Lyro API.
• Personal API keys authenticate your MCP client without re-authenticating every session.

Note: A full key is shown only once at creation time. Copy it immediately and store it somewhere safe - you cannot view it again.

For endpoints, authentication, and usage details, see API and developers.

Saved Replies

Saved Replies (canned responses) are reusable message templates for your team.

• Give each reply a title and body, and optionally a shortcut to insert it quickly.
• Use placeholders like {{first_name}}, {{name}}, or {{email}} - they resolve from the customer when the reply is inserted.

Properties

Properties are custom contact attributes that extend the data Lyro stores about each person.

• Add a property with a label and a key (the key auto-fills from the label).
• Choose a data type: text, number, boolean, date, single select, or multi select.
• Select and multi-select types require at least one option.

Properties show up on contact records in People.

Intents

Intents are your conversation taxonomy - the set of labels Lyro's classifier picks from when categorizing each message.

• Create intents with a label, a permanent slug, and a description.
• The slug is locked once created because it is stamped onto historical analytics events.
• Reserved fallback intents act as the classifier's safety net and are pinned to the top of the list.

A clean intent set keeps your reporting meaningful in Analytics.

Formatting

The Formatting tab holds message formatting presets - named rules that control how your agents structure and style their replies.

• Create a preset with a name and formatting instructions.
• Apply presets to keep tone and layout consistent across conversations.

AI CSAT

AI CSAT lets Lyro estimate customer satisfaction for conversations using a 1-to-5 rubric you control.

• Each tier (1 through 5) has a description that tells the evaluator what that score means.
• Edit the descriptions to match your standards, then save.
• Reset at any time to return to the default rubric.

Findings

The Findings tab turns on automated analysis of your conversations to surface gaps and improvement opportunities.

• Enable Findings to have Lyro analyze your recent conversations.
• Choose a mode: on-demand only, or a recurring nightly schedule plus on-demand.
• When nightly is selected, set the recurrence and timezone in the schedule editor.

Note: Findings works best when you have at least one knowledge base connected, so it can spot content gaps. Learn more in Findings.

MCP

The MCP tab connects Lyro to MCP-compatible clients and tools.

• Generate a personal API key to authenticate an MCP client.
• Paste the key directly into your MCP client configuration.

For the broader picture on connecting external systems, see Tools and integrations.