MailTwin
Setup

Set up MailTwin quickly

Three steps. Then you're writing email in your voice.

1. Install

  1. Download MailTwin.dmg.
  2. Open the disk image. Drag MailTwin into Applications.
  3. Open MailTwin from the Applications folder. The menubar will get a new ✉️ icon.

macOS will probably ask "are you sure you want to open this?" the first time. That's the standard "downloaded from the internet" warning — MailTwin is signed and notarized by Apple, so you're good. Click Open.

2. Pick your AI

On a recent Apple Silicon Mac (M1 or newer, macOS 26+), MailTwin will pre-select Apple Intelligence for you. It runs entirely on-device, costs nothing, and needs zero setup. Just hit Continue.

Want a more powerful model? Pick Claude, ChatGPT, Gemini, or any of the other four providers, paste your API key, and you're set. Not sure which to pick? Read the provider guide — three sentences will tell you.

3. Grant two permissions

macOS asks twice, because Mail and the system keyboard live in separate sandboxes:

Click Grant in onboarding and macOS will pop the right pane for each. Toggle MailTwin on, return to onboarding, and you're done.

That's it

Open Mail, click any message, hit ⌘⇧M from anywhere, pick Reply (my style). MailTwin reads what you typed last week, drafts a reply in the same voice, and streams it into the compose window. Tweak it like any other draft, hit Send.

Stuck somewhere?

Onboarding has a built-in self-test that catches the most common speed bumps (Mail not running, permissions not granted, no message selected). If you're past onboarding and something's broken, open Settings → App → Setup → Generate diagnostics report and email the result to [email protected]. We usually reply within a day, often the same hour.

Using the MailTwin workspace (build 74)

Open MailTwin Workspace from the menubar or the assistant footer. Today is the starting point; the sidebar opens each focused tool.

Inside a Mail draft: enable MailTwin Extension in Mail → Settings → Extensions and keep the MailTwin app running. Improve and My Style first hand off metadata for that exact compose session. Only after your button click does the host read or rewrite the resolved editor; ambiguous matches and failed validation leave the draft unchanged.

  • Today + Activity. See due and waiting items, replies confirmed by your last Refresh, quick actions, recent work, and local receipts labelled as success, failure, suggestion or learned choice.
  • Reply-aware Waiting On. Press Refresh in Workspace to run one bounded Sent/incoming-header metadata check. Replied requires exact RFC ancestry. Failure changes no status and triggers backoff; nothing runs automatically, in the background, or at launch.
  • Three Inbox Cleanup views. Clean Now stages a plan; Keep Clean lets you inspect or forget learned sender choices; Cleanup History shows receipts and exceptions. Nothing moves in the background.
  • Multilingual Voice Studio. Tune greeting, brevity, directness, warmth, never-use phrases and chosen examples separately for each account, with controls in all 15 languages.
  • Editable, reviewable compose. Edit generated text before inserting it. Improve Draft can show a red/green comparison, and rewrite validators still reject invented subjects, sign-offs and facts.
  • Structured action review. Edit title, owner, due date and Reminders/Calendar destination, deselect anything, then export the reviewed set once; duplicate lines are removed.
  • Cited Relationship Memory. Search the local Sent index, inspect source/date/recipient citations, open the exact message when its locator is available, and explicitly add at most three excerpts.
  • My Actions. Save reusable instructions with notes and optionally assign one unique ⌘⌥1–9 shortcut. The menubar or shortcut only opens the instruction visibly in the assistant; it never runs AI or sends mail automatically.

Sent-mail indexing is always your action. New installs default to 200 messages per account (up to 2,000). Index or Re-index runs locally only when you start it, in five-message Mail batches. Mail may be briefly busy; Stop prevents later batches but cannot interrupt the Apple Event already running. Nothing indexes automatically — not at launch, not on a timer, not in the background; the onboarding invitation and Voice Studio start it only when you press the button. After updating, re-index once only if older Relationship Memory results need exact source details or View in Mail.

Current provider defaults. Official catalogs checked 25 July 2026: Claude Opus 4.8 ($5/$25) and OpenAI GPT-5.4 ($2.50/$15) per million input/output tokens. Only an exact deprecated gpt-4o selection migrates; unrelated custom models stay untouched.

Nothing becomes silent automation. Learned cleanup choices are visibly proposed for review, Relationship Memory context is added only when you select it, and Replied appears only when exact message metadata proves a reply.