Codex edits your code and runs your commands inside an operating system sandbox you control. The CLI, the IDE extension, the desktop app and the cloud all run the same agent and read the same configuration: AGENTS.md, rules, skills, plugins, MCP servers and hooks.
Reach for it when
- You want an agent with an enforced boundary, not a promise
- You want the same setup in the terminal, the editor and CI
- You are already on ChatGPT or OpenAI models
Look elsewhere when
- You want a library to build your own agent on
- You cannot install anything locally
What the lessons will cover
- 01Install, and what you get
- 02The sandbox, demonstrated
- 03Permission profiles
- 04Approvals
- 05AGENTS.md and rules
- 06Prompting
- 07codex exec and JSON output
- 08Reviews, sessions and worktrees
- 09config.toml, skills and plugins
- 10MCP servers and hooks
- 11Subagents and the cloud
- 12A repository set up end to end
Others in coding agent
Common questions
Is Codex CLI free to use?
Codex CLI is open source and free to run yourself. You still pay whichever model provider you point it at, and this tutorial is free with no signup.
Do I need to know Python to use Codex CLI?
Basic Python is enough. Functions, dictionaries and imports cover most of what Codex CLI asks of you.
When should I not use Codex CLI?
You want a library to build your own agent on. You cannot install anything locally.
Lessons for Codex CLI are being written. Meanwhile the Agent Governance Toolkit tutorial covers the same ground: state, tools, loops, memory and human approval. Most of it carries straight over.
Start the Agent Governance Toolkit tutorial →