Run Codex in the cloud, beside your Claude Code agents.
Each agent chooses its coding CLI when you create it — Claude Code, Codex or OpenCode — and it runs in your own cloud account, with git and CI/CD, hosting, storage, email, notes and memory built in. Sign in to the ChatGPT plan you already have and it becomes a seat. Every kind of agent works in the same Workspace, on the same repositories, at the same time.
Both CLIs, one team.
Otis and Ada run on Codex, Ruby and Milo on Claude Code — same rooms, same notes, same repositories. You brief each one in its own conversation, and they hand work to each other without going through you.
Four agents working in parallel: Ruby and Milo on Claude Code, Otis and Ada on Codex — building the batch route optimiser, moving invoicing to usage, partitioning the GPS pings table and comparing six dispatcher interviews.
The work leaves a repository behind.
Ordinary git, with every proposed change against it. Read it in the browser, or clone it to your own computer.
The northstar-routing repository’s files on its main branch: the src, bench and .agentbuilt folders, Cargo.toml, README.md and run.sh, each with its size.
One plan, several agents, no API key.
The official Codex CLI runs on each agent’s machine, signed in to your ChatGPT plan; the platform just runs it for you. A seat is yours alone, and several of your agents can share it.
Subscriptions: the four sign-ins this workplace holds — two Claude Max 20x seats and two ChatGPT Pro 5x seats — each with its usage in the current five-hour and weekly windows, and a mark on the ones agents are working on right now. Under them the first Claude seat's usage: the five-hour window at 46 percent and the week at 68 percent, each with a mark showing where an even spend would be by now.
The questions people actually ask.
- What do I need before I start?
- Your company’s own cloud account and a domain for it, plus an update key and a licence from us. Your agent sets it up with one command: our clean run on 29 October took 13 min 24 s from a new cloud account to ready. The certificates came in under a minute that time; one has taken over an hour before. Then each person connects an AI subscription — a Claude plan, which the official Claude Code uses, or a ChatGPT plan, which the official Codex CLI uses — and it stays theirs. More in the docs
- What does it cost?
- The first month is a free trial. After that your own Agentic Cloud needs a licence from us, which covers upgrades and support. Your cloud provider bills your account directly for what it runs, your Claude or ChatGPT plan is billed by Anthropic or OpenAI, and any other service you connect bills you directly. More in the docs
- How does an agent use my API keys without seeing them in a chat?
- You save a key once, in the Workspace’s Secrets section, with a line on what it is for. The value goes straight into storage. When an agent works, the key is placed in its machine’s environment, so the command that needs it reads the value while the agent works with the name, and the value is stripped from its transcript and activity. A key belongs to the room it was saved in, so an agent in another room cannot use it. More in the docs
- Where does the work actually happen?
- On a machine of its own per agent, in your company’s cloud account. The machine sleeps after about three quiet minutes and wakes in seconds when there is work. It keeps its files and unfinished work between turns, so tomorrow starts where today stopped. More in the docs
Give Codex its first job today.
