Skip to content
ANYONE WITH A LAPTOP — YOUR PHONE, YOUR VOICE

A stream of thought becomes
a stream of product.

Ghost Studio gives anyone with a laptop voice command over it — from a phone. Keep the stream running — a stream of thought becomes a stream of tasks, and the stream becomes product. Large codebases, large projects, large anything: say what you want to build, change, or test — and the model to do it with; the agent configured on your own machine does the work. The moment you stop talking, that thought is already a task on its way to your machine. Your computer is the root node. Your phone is the remote.

Code any project you configure on your laptop — with just your voice. Your entire codebase is in play: the agent finds, reads, and edits whatever files the work touches, so you never have to name one. Features, fixes, refactors, tests: pair your iPhone, say what to build, and let your local coding agent edit files and run tests on your supported laptop or desktop — on whatever model you pick. There’s no send button to hunt for: the stream reads your natural pauses and ships each chunk of speech as its own task, so building starts seconds after the words leave your mouth. And it’s not just code: open a page, send an email, move files, launch an app — use your voice to do almost anything on your laptop. Root-node access means your computer does the work, not unrestricted root or desktop access. On-device transcription included.

  1. You keep speaking
  2. Your laptop keeps working
  3. Product ships
No App Store wait — home-screen app Your laptop or desktop Not just code — control your whole laptop Works on the lock screen Any model — one dropdown $19.99/month · cancel anytime
THE COMPANION WORKFLOW

Watch the product demo.

See a stream of thought become a stream of product — pair once, then talk your laptop through building a project, from recording and model selection to tasks, replies, and quota.

The real Ghost Studio companion, end to end — pair your phone with one QR scan, pick a model, and talk. Record a whole idea in one take, or keep the stream open — every pause naturally cuts the next chunk of your voice into a task and sends it. Whatever you say becomes a coding task on your laptop: describe a project and your machine builds it, then replies land back on your phone while you keep talking.

Download video
START CURIOUS. STAY IN CONTROL.

Speak the change.
Own the result.

Get the app, then use your account for billing and downloads.

ANNUAL PLAN

For the whole year

$190per year — billed once, cancel anytime

Same app, same control — one payment covers the year.

Get the app
  • Everything in the monthly plan
  • About 2½ months free vs paying monthly
  • Secure Stripe checkout
  • Account and subscription management
ON THE ROADMAP

For the whole team

Togethershaped by your workflow

Make room for more people, not more tabs.

Register team interest
  • Planned: shared projects
  • Planned: team permissions
  • Planned: shared usage controls

Plans are $19.99 per month or $190 per year; the account page shows exact prices and billing periods before purchase.

Your voice is the stream. Your phone is the interface. Your machine does the development.
A phone is all you carry
Browser-first workspace
A dedicated machine hosts your code
NO COMMANDS TO MEMORIZE

Say it like you’d
say it to a teammate.

Plain sentences are the whole interface — no syntax, no magic words. Whatever you say next is the next instruction: describe the change, the bug, or the idea — even the model to run it on — and your words become the task your laptop builds from.

  • “Fix the failing login test.”A bug report in one breath — the transcript is the task.
  • “Make the background black.”Design tweaks are as sayable as features.
  • “Refactor the checkout flow, then run the tests.”Multi-step asks fit in a single sentence.
  • “Put the strongest model on this migration.”Even engine steering is something you say.
NOT JUST CODE

Your voice runs
your whole laptop.

Coding is only the start. The agent on your Mac can do almost anything you would do at the keyboard — so you can use your voice to do almost anything on your laptop. Open a page, send an email, find a file, tidy a folder, launch an app: say it from your phone, and it gets done on your computer.

  • “Open my analytics dashboard in Chrome.”Pages, apps, and folders open on the Mac when you say so.
  • “Email Sam that the build is live.”Your Mac’s Mail app sends it from your own account.
  • “Move this week’s screenshots into a new folder.”Everyday computer chores, not just code.
  • Your rules still apply.It works through the agent you sign in, with the permissions you give it on your Mac — macOS may ask once before an app can be controlled.
TRANSCRIPTION, FRONT AND CENTER

Every transcript
becomes code.

Record an entire meeting, a design discussion, a lecture, or a long dictation — or send a transcript you already have, from anywhere, at any time. The voice doesn’t even have to be yours: whoever is speaking, the whole text, end to end, becomes the development task your machine builds from. Keep the stream running — thought in, product out.

  • Begin recording. Stop. Send.One tap captures the whole session — the full transcript arrives as a single task: build what was discussed.
  • Any transcript counts.Meeting notes, interviews, voice memos — send the text from anywhere and it works the same way.
  • Transcribed on-device.Speech becomes text locally, so the transcription stream stays free while it stays accurate.
NO SEND BUTTON — THE STREAM KNOWS

You stop talking.
It starts building.

The stream listens for the end of a thought, not a tap. Wherever you naturally pause, it cuts the chunk of voice you just spoke into a complete task and sends it to your laptop — while whatever you say next is still being captured. Code starts building seconds after the words leave your mouth, and replies land back on your phone while you’re still mid-flow.

  • A pause is the send key.Stop for a beat and the thought you just finished is already on its way — nothing to tap, nothing to re-say.
  • Natural chunks, whole tasks.The stream splits your speech where your ideas split — each piece arrives as its own complete instruction, exactly as you said it.
  • Fast by design.Tasks never queue behind your voice — several can be in flight at once, so a stream of ideas becomes a stream of finished work.
EVERY MODEL, ONE DROPDOWN

Pick the brain
behind the build.

No house model, no shortlist, no tier gate. Open the model dropdown and the coding engine on your laptop hands you its whole catalog — hundreds of models from every major lab, all in one list. Code with almost any model out in the world: tap a frontier brain for the hard refactor, a lightning-fast one for the quick fix, then switch again on the very next task — per task, not per plan. Rather keep talking? Say the model’s name and the instruction carries your choice.

The dropdown lists whatever your connected engine and provider account actually offer — picking a name never grants access or bypasses quota.

NO ONE HAS TO BE AT THE KEYBOARD

Locked screen?
Still building.

Your laptop doesn’t need you logged in at it — lock the screen, let the display go dark, and walk away. The Ghost Studio Mac app holds the machine awake while it runs, so every task you speak keeps landing, building, and replying on the lock screen. The only ask is power: keep it on with the lid open, and it works around the clock.

  • Lock it and walk away.The lock screen doesn’t pause anything — you never sit back down to unlock it for a task to arrive.
  • No babysitting the display.The screen can go dark — the Mac app keeps the computer awake while it runs instead of letting it drift to sleep mid-task.
  • Honest limit.A closed lid can still sleep the machine, and a powered-off computer can’t receive work — keep it on with the lid open.
IPHONE SETUP · SCAN ONE QR CODE

Your whole laptop,
minus the desk.

Pairing is one QR code — your account page shows it, your iPhone scans it, and you are connected to your Mac. No license key, no address to type.

01

Prepare your laptop.

Download Ghost Studio for Mac from your account page and open it — click Link this Mac in your browser and it joins your account. Keep your coding engine (Devin, Claude Code, or Cursor) signed in with your own account.

Locked screen is fine — just keep it on
02

Scan the QR code.

Tap Show QR code on your account page and point your iPhone Camera at it. Allow the download, then open Settings → Profile Downloaded → Install — Ghost Studio lands on your Home Screen, already connected to your Mac. No App Store download.

One scan — nothing to copy
03

Open it, then speak.

Open Ghost Studio from your Home Screen, allow the microphone, and try one small instruction. Your phone stays connected — no re-scanning.

Connected phones remember — no re-scanning

Not connected yet? Buying does not connect your phone by itself — open the Ghost Studio Mac app once, then scan the QR code on your account page. An active subscription and a coding engine signed in on your Mac are required.

Check availability
ROOT-NODE CONTROL, FROM YOUR PHONE

Your computer.
Your command center.

A serious coding setup with voice control, speech-focused transcription, and your own laptop doing the development work. Speak from your phone; the configured local coding engine edits files, runs tests, and iterates on the model you choose — with the permissions you grant.

Your computer. Your root node.

Bring your own supported laptop or desktop into the workflow. Use your voice to direct the local agent to edit files, run tests, and work with your tools — within your OS permissions. No hosted workspace required.

Voice command · configured local engine

Your whole codebase. By voice.

Point it at an entire codebase — not a snippet, not a single file. The coding agent on your laptop works across the whole project: it finds the right files, reads how they connect, edits across many of them in one task, and runs your tests. You talk about what you want; it handles where it lives in the code. It is the most natural way to build a project — just say it.

Full-codebase context · your local agent

Not just code. Your whole laptop.

Use your voice to do almost anything on your computer — open a page, send an email, find or move files, launch an app, run a script. Say it from your phone and the agent on your Mac does it, the same way it writes code. Your laptop, controlled by your voice.

Voice control · anything your agent can do

Away from your desk. Still in control.

Give instructions from your phone while your laptop stays running — lock the screen and walk away, and tasks keep landing on the lock screen. Speak a fix, a feature, the next test to run — even the model to run it on — and your authorized local coding engine handles the keyboard work. A working connection and completed pairing are required.

Voice development · your own laptop

Any model. Per task.

One dropdown holds your engine’s whole catalog — pick literally any model it offers, or just say its name out loud, and the task runs on it. Send your engine’s best to the hard problem and a faster model to the quick fix. What you can pick follows your connected engine and provider account.

Model choice · one dropdown

Record the whole meeting.

One tap on your phone starts a session recording — an entire meeting, a design discussion, a lecture, a long dictation. Stop, and the full transcript is delivered to your codebase as a single task: build what was discussed.

Session recording · phone app

Any transcript becomes code.

Already have a transcript? Send it from anywhere, at any time — the entire text, end to end, becomes the instruction your machine works from. No re-typing, no summarizing.

Transcript → code · phone app

Speech-focused transcription.

The local transcription path uses voice activity detection to skip non-speech audio. Speaking in a busy room or restaurant? Keep the microphone close and check the transcript — nearby voices can still be picked up.

Local speech filtering · not speaker isolation

Review before you release.

A planned space to inspect changes and verification results before approval. Review work on your own machine before publishing.

Review flow · product direction

You stop. It sends.

No send button to find — the stream naturally cuts your speech into tasks wherever a thought ends. The moment you pause, that chunk is already on its way to your laptop, and several tasks can be in flight at once while you keep talking.

Pause = send · parallel tasks
Where we are today

The phone app works with the Ghost Studio Mac app on your own Mac; this website does not run coding agents. You can buy and start today — the phone side is a home-screen web app, so there is no App Store wait. A native iOS client is in development and in Apple’s review. App Store approval is a goal, not a guarantee. Hosted workspaces are still in development. Check your account for release and purchase availability.

Help shape the launch
A FEW GOOD QUESTIONS

Glad you asked.

Clear about what’s here.
Honest about what’s next.

Who is Ghost Studio for?

Anyone with a laptop who wants to command and control it from a phone. Point your voice at large codebases, large projects — large anything — and the coding engine configured on your own machine turns spoken instructions into real work: features, fixes, refactors, tests, and reviews — while you are away from your desk. You do not have to be a developer: plain sentences are the whole interface. It is not a hosted IDE or a no-code app builder — you still work in real projects, on your own machine, with the engine and models you configure.

What can I actually say?

Whatever you would type — plain sentences, no commands or syntax to memorize. “Fix the failing login test,” “make the background black,” “refactor checkout and run the tests” all work the same way: what you say is transcribed and becomes the task your configured coding engine builds from. Steering is spoken too — say which model should run a task, or send a transcript you already have. The stream keeps your earlier speech as context, so you can think out loud instead of composing perfect prompts.

Can it work with my entire codebase?

Yes — you point your voice at the whole project, not one file at a time. The coding agent configured on your laptop opens your actual repository, searches it, reads the files it needs, follows how they connect, and edits and tests across as many files as the task touches — so you can say “rename the user model everywhere and fix whatever breaks” and it manages the files for you. You never have to say which file to open. How much it holds in mind at once follows the engine and model you connect, and it works within the project folders and OS permissions you give it.

Is it only for coding?

No — coding is just the start. Because the agent on your Mac can run anything you could do at the keyboard, your voice can drive the whole laptop: “open the Stripe dashboard,” “email Sam that the build is live,” “move today’s downloads into a folder,” “start the music.” Pages open in your browser, email goes out through your Mac’s Mail app from your own account, and files and apps are handled right on the computer. It works through the coding agent you sign in, within the permissions you give it — macOS may ask once before the agent can control an app like Mail, and anything that needs a human check (a CAPTCHA, a password prompt, a two-factor code) still waits for you.

Can I use it in a busy room or restaurant?

Yes — it works in a busy room or restaurant: keep the phone close and speak at a normal volume, and the stream follows the voice nearest the mic. The local transcription path uses voice activity detection to skip non-speech, so background ambience is not the blocker. One honest limit: it is voice detection, not speaker isolation — we cannot guarantee only your voice reaches the transcript, so a nearby conversation louder than you can bleed through. Glance at the transcript when it is loud around you.

What does root-node control mean?

Your own computer is the execution hub; your phone sends instructions to its configured coding engine. Use a supported laptop or desktop you own and authorize. Through that agent your voice can do almost anything on the laptop — code, open pages, send email, manage files — but it is not screen-sharing remote desktop and not a grant of administrator/root privileges. Agents inherit the local process’s OS permissions; choosing a project is not an OS sandbox. Compatibility depends on your operating system and coding engine, and pairing and license setup are required.

Do I still need a laptop or computer?

Yes — your phone is the remote; your Mac runs the Ghost Studio Mac app (downloaded from your account page) and your local coding engine. You can be away from it, its display can be off, and it can sit on the lock screen — nobody has to be at the keyboard — but the computer must remain powered on, awake, and online. The Ghost Studio Mac app holds the machine awake while it runs, so you are not babysitting the display. A sleeping or shut-down computer cannot receive or run local tasks. The HTTPS connection only carries instructions; it does not keep the computer running or grant administrator/root privileges. A hosted workspace we operate is still on the roadmap.

Does my laptop have to stay unlocked?

No — the lock screen does not stop the work. Lock the screen, let the display go dark, and walk away: your spoken tasks keep landing, building, and replying while nobody is at the keyboard. You never have to sit back down and unlock it for a task to arrive. The honest limit is power, not presence: a closed lid can still put the machine to sleep, and a sleeping or powered-off computer cannot take work — keep it on with the lid open and it keeps building around the clock.

Do I need the App Store?

No — you can start the minute you buy. The current phone companion is an installable progressive web app (PWA), so there is nothing to download from the App Store and no review to wait on: tap Show QR code on your account page and scan it with your iPhone camera, allow the Ghost Studio profile (Settings → Profile Downloaded → Install), and it opens full-screen from its own home-screen icon like any other app. A native iOS client is in development and has been submitted to Apple for review; it is not on the App Store yet, and you do not need it to use Ghost Studio. Its workflow requires explicit consent to trust and pair your phone with your laptop, and microphone permission for user-started voice capture. The phone sends your instructions; coding tasks run on your laptop, not on your iPhone.

How does my phone connect to my laptop?

Through your Ghost Studio account — no license key, no address, no port forwarding. Open the Ghost Studio Mac app once and click Link this Mac in your browser; then tap Show QR code on your account page and scan it with your iPhone. Both devices sign in to the same paid account and exchange tasks through ghoststudio.app over HTTPS, so they work on different networks and your phone reconnects on its own. An active subscription is required.

Do I have to install Devin or an editor?

The local workflow needs a supported coding engine configured and authenticated on your own machine, such as Devin Local or an IDE integration. That engine works within the project access you authorize. A hosted experience without local installs is still on the roadmap.

Is the hosted coding service available now?

No — hosted coding is not generally available. The Ghost Studio Mac app runs tasks on your own running Mac; this website does not provide hosted coding workers. Subscriptions are sold through Stripe checkout on the account page.

Can I choose any model?

Yes — open the model dropdown and choose from the connected engine’s entire catalog: hundreds of models from every major lab can sit in that one list, so you can code with almost any model out there. You can also just say which model to use — the instruction carries your choice to the engine. Switch as often as you like — the next task can run on a different model than the last. Model choices still depend on that engine and your provider account; execution requires the exact model to be supported and available on your account, and choosing a name does not grant model access or bypass quotas.

How good is the coding?

As good as the engine and model you connect — and you can use any model your engine offers, picked from the dropdown or named out loud per task. Ghost Studio is the steering, not the engine — point it at a capable coding engine and a frontier model on your own laptop, and spoken instructions become real development work: features, fixes, refactors, and tests inside your actual project. Your engine and model choices — not a plan tier — set the ceiling.

Is Devin just a dependency you can bundle?

A CLI integration does not automatically grant redistribution or resale rights. Any commercial Devin integration needs appropriate provider terms, authentication, usage accounting, and quota handling. Ghost Studio does not bundle Devin or share a personal subscription with customers.

How does it know when to send?

You don't press send. The stream listens continuously and treats a pause as the end of a thought — the chunk of speech you just finished is sent to your laptop as its own task while whatever you say next is still being captured. Start a new idea where the last one ended and each piece lands as a complete instruction; long one-take recordings work too, shipping as a single task.

Does the voice have to be mine?

No — the instruction comes from whatever the microphone hears: your voice, a teammate's, or a whole group in a meeting or design discussion. Hand the phone around or set it in the middle of the room; the transcript becomes the task your configured laptop builds from, on whatever coding engine and model you connected.

What will a subscription cost?

$19.99 per month on the monthly plan, or $190 per year on the annual plan, billed through Stripe — cancel anytime. The account page shows the exact configured price and billing period before you pay. Provider model usage is separate unless a plan explicitly includes it.

Will I have control over my code?

That is the intended hosted design: explicitly selected repositories, isolated workspaces, visible model and usage information, and a review step before changes are published. These safeguards still need implementation and verification before live hosted execution launches.

How do I get help?

Email officialghoststudios@gmail.com — for setup, pairing your iPhone and Mac, signing in your coding agent, billing, or any question before or after you buy. The support page also has step-by-step setup and fixes for common problems.

NEED HELP?

Questions or stuck on setup? Email us.

This inbox is for any help with Ghost Studio: installing the Mac app, pairing your iPhone, signing in your coding agent, billing, or a question before you buy.

officialghoststudios@gmail.com

Prefer to fix it yourself? See the setup and troubleshooting guide.

NOTHING BETWEEN YOUR VOICE AND YOUR CODE

Say it.
Ship it.

Keep the stream running — every thought you speak becomes product, on the model you choose, even while your laptop sits locked.

Get it now — start today