AGI Lab
...

Voice Control for macOS

AGI Controller

A voice controller for macOS that is always ready, acts before you finish the sentence, and does not guess.

How it works

What you say × what happens

Say it in English or Japanese. Examples:

"open Safari"

「Safariを開いて」

Open an app or website

"click log in"

「送信を押して」

Press a button or link on screen

"type hello world into the search field"

「こんにちはと入力して」

Type exactly what you said

"search for alan turing"

「東京の天気を調べて」

Search the web

"scroll down a bit"

「一番下まで」

Scroll

"put this on the left half"

「左半分にして」

Arrange windows

"export as PDF"

「サイドバーを表示して」

Choose a menu bar command

"run my morning routine"

「ショートカットの日報を実行」

Run a macOS Shortcut

"no, not that one"

「違う、そっちじゃない」

Reverse the last action
On and off

No key to press

There is no key to hold. Listening is a switch, on or off, and while it is on you just speak.

Turn it off

"stop listening"

Closes the microphone. Turn it back on from the menu bar icon.

Pause it

"pause for five minutes"

One, five, fifteen minutes or an hour. It resumes by itself.

Call it by name

"hey controller, …"

An utterance with the name is always taken as a command. The names are a setting.

Checks

It does not guess

When it is not sure, it checks before it acts.

Ambiguous target

Numbered badges appear on the candidates. Say "two" to pick one.

Destructive actions

Send, delete, buy, discard, and quit wait for a spoken "confirm".

Wrong action

"No, not that one" reverses the last action.

Speech around you

It first judges whether you are talking to it. Conversation, thinking aloud and video audio get no reaction and show nothing. It pauses by itself during calls.

How it works

🎙️

Just speak

While listening is on, there is no key to press.

📝

Transcribed on your Mac

Apple on-device speech recognition.

🧠

JEV answers with probabilities

About fifteen questions in a few hundred milliseconds.

⚡

Code decides

Act, wait, ask, confirm, or ignore.

A decision on every transcript change

Every time the partial transcript changes, one request goes to JEV, the System One model from TypeSafe. JEV answers about fifteen typed questions with calibrated probabilities in a few hundred milliseconds. Code decides what happens next: act, wait, ask which one, ask for confirmation, or ignore.

Is this addressed to me?Is the command complete?Which action?Which app?Which control on screen?Is it destructive?

JEV never generates text

What to type, which URL, which app, and which on-screen control are candidates extracted by code. JEV only picks one, and the pick is copied verbatim. No generative AI is involved, so nothing you say is rephrased.

No text generationPicks from candidatesCopied verbatim

Only text leaves your Mac. Audio never does.

Speech is transcribed on your Mac with Apple on-device speech recognition, and no screenshots are taken. Requests go to AGI Backend with your sign-in and from there to JEV. The app holds no API key.

What leaves the Mac (text only)

  • The text of what the microphone heard, for the "is this for me?" check (not sent in "only when I say its name" mode)
  • The name, window title, and URL of the frontmost app
  • Only for utterances judged to be commands: up to 100 control labels and 150 menu command names
  • Names of running, Dock, and mentioned apps
  • Your Shortcut names
  • The last three actions
  • With AGI Cockpit installed: its task names and open choices

What never leaves the Mac

  • Audio
  • Screenshots
  • The content of password fields
  • Your clipboard
  • Usage insights
On-device speech recognitionNo audio uploadNo screenshotsNo API key in the app

Features

It reads the frontmost window through the macOS Accessibility API, so it works across apps.

Works across apps

Checked with Finder, Safari, Chrome, System Settings, and Electron apps such as Discord. It reads up to 100 control labels from the frontmost window.

Notch HUD

Live transcript, intent and target probability bars, numbered candidates, latency, and estimated cost. The probability bars can be hidden.

Dictation mode

Say "start dictation" and every sentence is written word for word until you say "stop dictation". Nothing is rephrased.

English and Japanese

Both the UI and speech. Say "switch to Japanese" to change the listening language.

Shortcuts and menu commands

Runs your macOS Shortcuts by name and chooses menu bar commands.

Windows, scrolling, and keys

Left half, right half, maximize, minimize. It also scrolls and presses keys from a closed set.

AGI Cockpit

With AGI Cockpit installed, say "stop the current task" or "go with option B".

Automatic updates

Updates are delivered through Sparkle.

AGI Lab login

Sign in with an AGI Lab account that has an active membership.

Requirements

System

  • macOS 26 (Tahoe) or later
  • Apple Silicon
  • An AGI Lab account with an active membership

Permissions

  • Accessibility
  • Microphone
  • Speech Recognition

Download

macOS 26 (Tahoe) or later / Apple Silicon / AGI Lab account with an active membership required

AGI Controller - A Voice Controller for macOS