The tone row sits ABOVE the smart-reply chips so you choose tone first, then click a chip to fire generation. The row is a single horizontal scroller at every screen width, on desktop as well as on a phone, and it never wraps to a second line. Your own default tone is always the first pill. Press Alt+T to cycle through tones from the keyboard.

• Customer-friendly — the new default. Warm but professional, like a thoughtful customer-success rep.
• Match my style — uses your personal writing-style profile so replies mirror your voice (sentence length, formality, sign-offs).
• Match recipient — mirrors the recipient's style based on their prior messages in the thread.
• Brand — uses the org brand-voice description set by an admin. Hidden when no brand voice is configured.
• Match sender, Warm, Professional, Direct, Formal, Casual, Brief — concrete tone presets.

Picking a tone on the strip applies to that one message. To change the voice Laureo uses from now on, open Email settings (the gear in the inbox toolbar, or Settings > My preferences > Email inbox), go to AI in your inbox, and set Default reply tone. That one picker governs every reply Laureo writes for you, including auto-drafts, unless you pick a different tone for a particular message.

Until you choose one, the row's description tells you which default you are actually inheriting, the workspace default or the app default, rather than showing an inherited value as though you had picked it.

Under AI in your inbox in Email settings there are two scope pickers. Both are per user: they govern your own mailbox and change nothing for your teammates. Each offers the same three choices.

• Off: nothing runs automatically for that surface. On Auto-draft replies this genuinely stops the drafting. Nothing is queued, so no draft arrives.
• Emails from your contacts: only mail that Laureo has matched to a person or company in your CRM.
• All real conversations: the widest setting. It is deliberately not called “everything”, because newsletters, calendar invites, out-of-office replies and automated mail are always skipped no matter which scope you pick.

• Auto-draft replies controls whether Laureo prepares a reply in advance for incoming mail. It never sends one for you.
• Suggested actions controls which emails get the one-click AI action suggestions (reply, log, create task).

Once an AI draft is generated, a refinement bar appears with these one-click actions: Shorten, Elaborate, Polish, Formalize, Casual, Match my style. Each action streams a rewrite of the entire draft. Undo rolls back to the previous version (up to 3 turns of history).

Select a sentence or paragraph, then choose a rewrite action. The AI streams a replacement with the same meaning but a different shape — shorter, longer, more formal, more casual, or matched to your style. You can also type a free-form instruction.

Click the AI button in the compose toolbar, type a short instruction (“ask about pricing”), pick a tone, and the draft streams character-by-character. Recent activities and CRM context for the recipient are pulled in automatically.

Inside an inbound thread, click Auto-draft reply next to the inline- reply textarea. The AI streams a full draft using the recipient's engagement signal, recent activities, and the prior message in the thread — no instruction text from you. Edit before sending; the AI never auto-sends.

Each generated draft surfaces a Why this draft panel listing the CRM facts cited (open deals, recent quotes, latest meeting notes, engagement signal, ticket history). Click any cited item to open the underlying entity in a new tab. Items dropped because of token budget appear in a collapsed list below.

A conversation of two messages or more carries an AI summary available bar above the message. Open it and you get the key points first, then any action items, written as they arrive. Nothing is generated until you open it, so a thread you only glance at costs nothing. Once you have opened one summary in a browsing session, later threads are prepared just ahead of your click so the panel fills faster.

A single message gets the same bar once it is long enough to be worth summarising: about 1,200 characters of readable text, counted after quoted replies and forwarded copy are stripped out, so a two-line note under a long quoted history is not treated as long. The bar reads AI summary available (this email) in that case, and Summarize Thread in the message menu scrolls to the same card and follows the same rule.

The two universal chips — Quick thanks and Custom… — are pre-rendered when the email opens, so you can act in under 100 ms. Context-aware dynamic chips stream in alongside them as the LLM responds. Most inbound emails are warmed at sync time, so dynamic chips usually land in under 200 ms on the first open; a cold-cache open (rare) takes 1–2 seconds. Once chips exist for an email they are cached for 7 days and re-served instantly on subsequent opens.