Skip to content
WebAuditOS
Start free audit

Test your website now

Free, no card. We read only public pages.

Run free audit →

Vibe coding tools

What is OpenAI Codex?

Codex is OpenAI's coding agent for the terminal, IDE and cloud. What it does, which ChatGPT plans include it as of November 2026, and how to use it safely.

Codex is OpenAI's AI coding agent. It reads a codebase, writes and edits code, runs commands and tests, and works in your terminal through the Codex CLI, in your editor, in the ChatGPT app or as a cloud agent running tasks in parallel. As of November 2026, it is included in every ChatGPT plan, with usage limits by plan.

Published · Last updated

By Michal Měřínský. Audit methodology.

What Codex does

  • Codex CLI: an open-source agent that runs in your terminal on your machine.
  • IDE extension: the same agent inside VS Code and similar editors.
  • Cloud: tasks run in isolated environments in the cloud, several at once, and come back as changes or pull requests.
  • Integrations: on paid plans, code review on GitHub and requests from tools such as Slack. The name Codex was also used for an older OpenAI code model from 2021. Today it refers to this agent.

Which plans include it

OpenAI's documentation on 9 November 2026 lists Codex as included with ChatGPT Free, Go at 8 dollars a month, Plus at 20, Pro from 100, Business at 20 dollars per user billed annually, Edu and Enterprise. Codex shares usage limits with other ChatGPT work, and limits vary by plan and model. Codex can also run on an API key, billed at API token rates, without cloud features.

Using Codex safely

  • Start in a mode that asks for approval before running commands or editing files, and loosen it only for trusted repositories.
  • Commit before large tasks so you can review and revert the diff.
  • Keep secrets out of the working directory, or ignore them explicitly.
  • Review cloud task results as pull requests, with tests, before merging.
  • Write an AGENTS.md file describing your conventions, commands and security rules for the repository.

An agent that can run commands can also run destructive ones. Approval prompts exist for a reason.

Codex and the code it ships

Codex writes code that compiles and passes the tests it can see. It does not know your production configuration, your security headers or whether the deployed site renders correctly for crawlers. Check those after deploy.

What a scan can and cannot tell you

WebAuditOS checks the deployed site Codex helped build: security headers, console errors, duplicate titles and error pages that leak stack traces.

It cannot review Codex's diffs or your repository.

Ask the agent to fix scan findings by pasting the evidence into the task. Then rescan to confirm.

Short answers

Is Codex free? It is included in ChatGPT Free with limited usage, and in paid plans with more.

Is Codex the same as ChatGPT? It is OpenAI's coding agent, available through ChatGPT plans and as a CLI.

Is the Codex CLI open source? Yes.

Codex or Claude Code? Both are coding agents. See our comparison for the differences.

In the product

The checks this maps to

Every check named here runs on every scan, on every plan - what a plan changes is how many pages are crawled, how long the history is kept and what you can do with the result. These are the ids the findings carry, so you can match a line in a report to the explanation above.

Vibe coding tools

Codex vs Claude Code: which should you use?

OpenAI's Codex and Anthropic's Claude Code compared in November 2026: where they run, plans and pricing, how they handle permissions, and how to choose.

Vibe coding tools

What is Claude Code?

Claude Code is Anthropic's coding agent for the terminal, IDE, desktop and web. What it does, which plans include it in November 2026, and how to use it safely.