AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: My Claude And Codex Workflow Across Several Macs With T3 Code on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get audio and creator gear delivered free with Prime

  • Fast, free delivery on millions of items
  • Prime Video, Amazon Music and more included
  • Member-only deals all year
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

TL;DR

Thorsten Meyer says he now uses T3 Code to manage Claude Code and Codex across eight Macs and a Linux server from one workspace. The open-source app connects to agents signed in through users’ existing provider accounts; Meyer says he has not measured a productivity gain, and some multi-machine features remain in beta.

Thorsten Meyer says he has moved his Claude Code and Codex sessions on eight Macs and a Linux server into one T3 Code workspace on his laptop. The change puts the AI coding agents he uses to build and check software behind a shared interface, rather than separate terminal windows on separate machines; Meyer says he has not measured a productivity gain.

In an account published November 18 on ThorstenMeyerAI.com, Meyer said each Mac runs Claude Code and Codex, with a Linux server also connected. He describes T3 Code as the shared workspace through which he follows sessions across those machines. The project’s makers call the app a “control plane for coding agents.” It connects to agents users already run and sign in to, rather than reselling their model tokens, according to the project’s materials.

Meyer’s earlier setup split Claude Code between terminal windows, a terminal multiplexer and the Claude Code desktop app. He used Codex in other terminal windows to check work. With sessions distributed across machines, he said, he had to locate the relevant computer to find out what was running or waiting for approval. Passing a change from one agent to another also meant carrying context between tools himself.

He says his current workflow uses one agent to build and another to review, with T3 Code providing a shared thread list. The project documentation describes several ways to connect machines, including a local network, SSH and Tailscale HTTPS. Its optional T3 Connect service is described as free “for now,” with a limit of three machines per account; the source also names other connection methods and a paid plan for connecting more. Meyer says he has not quantified whether the arrangement makes him more productive.

At a glance
reportWhen: Published November 18, 2026; T3 Code ve…
The developmentThorsten Meyer has described moving his multi-machine Claude Code and Codex workflow into a single T3 Code workspace.

My switch to T3 Code · before and after

From terminal windows on many Macs to one workspace

The agents did not change. Claude Code and Codex still run on the Macs that do the work. What changed is where I see them: in one list of threads, on the laptop or the phone.

Before After Who did what? Who waits? Needs you Claude builds Codex checks Working Phone = remote
  1. 1 · BeforeSessions sit in terminal windows on each Mac. Nothing tells you which agent did what, or which one waits for you.
  2. 2 · One workspaceEach Mac runs T3 Code. One app connects to all of them, over the local network, Tailscale, SSH or T3 Connect.
  3. 3 · ThreadsOne thread per task. Claude Code builds, Codex checks. Busy threads fold away; a thread that needs an answer comes back to the top.
  4. 4 · RemoteThe phone approves the next step. The Mac keeps doing the work and has to stay on.

Sources: T3 Code README and user docs (remote access, working with threads, mobile notifications, FAQ), read November 18, 2026. Setup: the author’s own. Drawing: a generic interface, not a product screenshot.

Graphic: Powered by Thorsten Meyer AI

One View Across Coding Machines

The development speaks to a practical problem for developers running multiple coding agents in parallel: keeping track of work and requests for approval across different computers. A shared interface can reduce the need to check each machine separately, according to Meyer’s account, and place build and review sessions closer together in one workflow.

That does not establish a measurable performance benefit. Meyer explicitly says he has no productivity figure to report. The more immediate change he describes is organizational: sessions are easier to locate, and an agent waiting for input can be surfaced in the same workspace as other work. For readers using just one agent on one computer, he says, there may be little reason to switch.

Products Worth Considering

DJI Mini 4K Fly More Combo Drone with 4K Camera and 3 Batteries

DJI Mini 4K Fly More Combo Drone with 4K Camera and 3 Batteries

Apple MacBook Air 13-inch with M5 Chip, 16GB Memory, 512GB SSD, Midnight

Apple MacBook Air 13-inch with M5 Chip, 16GB Memory, 512GB SSD, Midnight

As an affiliate, we earn on qualifying purchases.

From Separate Terminals to Threads

Meyer’s use case depends on coordinating agents that have different jobs. Claude Code handles some implementation work, while he uses Codex to check changes. Before using T3 Code, those roles were spread across separate terminal sessions and applications. He says the difficult part was not that either tool failed on its own, but that the handoff and monitoring across tools and machines required manual attention.

T3 Code is described in the source as a free, open-source app under the MIT license. Its documentation lists Claude Code, Codex and several other coding agents. It also describes Git worktrees, which can give threads separate working copies, and an option to start another thread in the worktree containing an existing change. These are project-documented capabilities; Meyer’s account does not provide an independent evaluation of every feature.

The source identifies version 0.1.10, dated November 17, 2026, and characterizes the project as young and fast-moving. It says several multi-machine features are still in beta. Phone and browser access can be used as remote controls, but the documentation says the computer doing the work must remain on.

““I have not measured a productivity gain, so I won’t quote one.””

— Thorsten Meyer, writing on ThorstenMeyerAI.com

Beta Features and Practical Limits

Meyer’s account is a first-person description, not a comparative test. It does not report measured time savings, error rates, or results from other users. It also does not establish how reliably the workflow performs across different networks or machine configurations.

Some multi-machine features remain in beta, according to the source, and T3 Connect is described as free “for now.” The source says that service supports up to three machines per account and identifies a paid plan and alternative connection routes, but pricing and future service terms are not detailed in the account. The project’s version number and rapid pace of development also mean that features or limits may change.

Users still need to follow the terms and usage limits of their own AI coding providers, Meyer warns. He also says new threads start in full-access mode; readers should check the app’s current permissions and settings before connecting it to projects or machines. The account does not specify how every security or access-control detail is handled.

More Machines, Same Provider Accounts

Meyer says additional machines can join his workspace, but he does not give a date for adding them or announce a formal product milestone. The immediate next step for the project is continued development of its machine-connection and agent-coordination features; the source offers no release schedule.

Readers considering the same approach can compare the current connection options and plan limits in T3 Code’s documentation, then check that their machines can remain available when remote sessions need them. Any decision to adopt it should account for the project’s beta status, provider terms and the user’s own requirements for permissions and remote access. Meyer’s account leaves open whether the workflow produces productivity gains beyond making sessions easier for him to monitor.

Key Questions

What change did Thorsten Meyer report?

He says he now manages Claude Code and Codex sessions on eight Macs and a Linux server through one T3 Code workspace on his laptop.

Does T3 Code provide Claude or Codex subscriptions?

No. According to the project materials quoted by Meyer, T3 Code connects to agents users have already signed in to and does not resell tokens. Users remain subject to their providers’ plans and terms.

Can a phone run the coding agents remotely?

The source describes phone apps as a remote interface for machines doing the work. The computer running the agents needs to stay on; the phone does not replace that machine.

Has Meyer shown that T3 Code improves productivity?

No measured productivity result is given. Meyer says he has not measured a gain and describes workflow benefits, such as finding sessions and approvals in one place, as his own experience.

What should users check before trying it?

Check current connection limits, beta features, provider terms and account usage limits. Meyer also says new threads start in full-access mode, so users should review permissions before connecting it to their machines and projects.

Source: ThorstenMeyerAI.com

You May Also Like

How AI-Assisted Review Connected The Dots In The PHP2Shell Story

A PHP security timeline links a 2004 warning, an old out-of-memory bug and 2026 AI-assisted research behind two remote-code-execution fixes.

Every Benchmark Launched 2023-2024 Has Fallen — The METR / SWE-Bench / CORE-Bench / MLE-Bench / PostTrainBench Sequence

Every major AI research benchmark launched in 2023-2024 has either saturated or is nearing saturation, signaling accelerated AI capability growth.

Google’s Gemini 4 Neon Breakdown: Performance, Price, And Caveats

Google’s Gemini 4 Neon scores one point above Carbon at xhigh and costs less per benchmark task, but comparisons and test results have limits.

On‑Prem Vs Cloud for Training: a TCO Framework

Understanding the TCO framework for on-premises versus cloud training helps you make informed decisions—discover which option best fits your long-term goals.