Skip to content

Codex Cloud

A tutorial — run a coding task in an isolated hosted environment, review the diff, then open a PR. Cloud is one Codex surface, not a second product. Sandbox, approval, and AGENTS.md still apply.

Official landing: learn.chatgpt.com/codex/cloud. Docs: Codex cloud. Entry: chatgpt.com/codex.

Prerequisites

NeedRequirement
AccountA ChatGPT plan that includes Codex. Limits live on the pricing page
GitHubA GitHub account you can authorize, plus the repos Codex should see
Browserchatgpt.com/codex
Optional CLIInstalled Codex CLI for codex cloud / codex cloud exec

Learning objectives: stand up one Cloud environment; start a task from the web or CLI; review the summary and diff; know when Cloud beats the laptop.

Non-goals: local sandbox setup (CLI); IDE first chat (IDE); security scans (Security); plan numbers (Plans).

What Cloud is — and is not

Cloud runs Codex on OpenAI-hosted machines against a reproducible environment you configure per repository. Each task gets its own isolated env. Several tasks can run at once. You review when they finish.

It is not:

  • A fourth product. Hosted code review / QA is a Cloud workflow.
  • A replacement for the laptop. Local CLI / IDE stay faster for a focused edit.
  • ChatGPT Work Cloud. Work Cloud finishes a deck or research job; Codex Cloud writes code and opens a PR.

Desktop Codex stays in the ChatGPT desktop app. That surface is mapped on the product line and in Work vs Codex — do not look for a second desktop tutorial here.

When to use Cloud

Official “use Cloud when…” list:

SituationWhy Cloud
The job should run in the backgroundThe laptop stays free
You want several attempts in parallel--attempts / multiple chats, then pick
Work starts in GitHub, Linear, or SlackOfficial integrations dispatch into Cloud
You are away from the development machineWeb UI or codex cloud exec

Stay local when you need this machine’s files, a debugger attached to the current process, or a 30-second edit on the file you already have open (IDE).

First 15 minutes

1. Open Codex and sign in

Go to chatgpt.com/codex and sign in with the same ChatGPT account you use for the CLI.

2. Connect GitHub

Authorize GitHub when prompted. Grant only the repositories Codex should see.

3. Create an environment

Open environment settings. Create an environment for the repo. Configure dependencies, tools, environment variables, and secrets the task needs.

Details: Cloud environments. Internet: Agent internet access.

4. Start a task

Back on chatgpt.com/codex, pick the environment and describe the result, not the vibe:

text
Add a failing test for the date parser in packages/core, then make it pass.
Do not touch packages/legacy. Open a draft PR when tests are green.

Watch the logs, or leave it running.

5. Review, then merge on your terms

Read the summary and the diff. Ask for a follow-up, or open a pull request when the work is ready. Do not merge on “looks plausible.”

From the CLI

bash
codex cloud                                            # browse environments (Ctrl+O reveals IDs)
codex cloud exec --env <ENV_ID> "run the migration dry run"
codex cloud exec --env <ENV_ID> --attempts 3 "..."     # 1-4 attempts

--attempts is for nondeterministic tasks. Pick the best result; do not average them.

Dispatch from GitHub, Linear, or Slack

Start a Cloud task without leaving the pull request, issue, channel, or thread.

FromOfficial page
GitHubUse Codex with GitHub
LinearUse Codex in Linear
SlackUse Codex in Slack

The IDE can also hand a long task to Cloud and return you to the same chat: delegate from the IDE.

Hosted review is Cloud, not a fourth product

SurfaceWhat it does
Local /reviewBase-branch or uncommitted diff; does not change the tree
Cloud code review / QAReview and QA in the hosted environment. Eligible customers get GPT-5.6 Sol; Cloud selects the model
Codex Security ReviewPR + repo context + threat model. Research preview; see Security

Local /review is a session command. Do not reuse the hosted-review name for it.

Common pitfalls

PitfallWhat happensDo this instead
Skipping the environmentMissing deps, failed installs, empty diffsConfigure the env before the first real task
Treating Cloud as a second laptopSecrets and local-only tools are not therePut secrets in the env; keep machine-local work local
Merging the first plausible PRTests never ran, or ran against the wrong treeRead the diff; require the test output
Confusing Cloud with Work CloudYou get a deck instead of a PRSwitch to Codex at chatgpt.com/codex
Expecting Cloud while the laptop sleeps for RemoteDifferent productsRemote drives a connected computer; Cloud does not need that machine

Real-world use

A frontend monorepo: three Cloud chats try the same flaky migration with --attempts 3. You keep editing the design-system package in the IDE. The winning Cloud attempt opens a draft PR. You review it on the phone with Remote only if the laptop is still the review host — otherwise stay on the web Cloud UI.

Next steps

  1. Environment knobs → Cloud environments
  2. Internet policy → Agent internet access
  3. Editor handoff → IDE
  4. Phone approvals on a local host → Remote
  5. Wiring CI on your runners → Project Integration

Official sources

Built for frontend engineers · Powered by VitePress