npm i -g @openai/codex
or brew install --cask codex

Codex CLI is a coding agent from OpenAI that runs locally on your computer.

If you want Codex in your code editor (VS Code, Cursor, Windsurf), install in your IDE.
If you are looking for the cloud-based agent from OpenAI, Codex Web, go to chatgpt.com/codex.

Running this fork locally

If you are sharing a fork and want others to run your exact branch:

# 1) Clone your fork and checkout your branch
git clone https://github.com/<your-username>/codex.git
cd codex
git checkout <your-branch>

# 2) Build the CLI
cd codex-rs
cargo build -p codex-cli

# 3) Run it
./target/debug/codex

If you want a dedicated command name for this fork, add a shell alias:

# from codex-rs/ inside this repo
echo "alias codexrlm='$(pwd)/target/debug/codex'" >> ~/.zshrc
source ~/.zshrc

# now run your fork build with:
codexrlm

If you want the full setup (toolchain, dependencies, and packaging details), see docs/install.md.

Native RLM settings (no env vars required)

On first run, Codex writes missing [native_rlm] defaults into ~/.codex/config.toml. You can then tweak Native RLM directly there:

[native_rlm]
enabled = true
max_iterations = 20
max_llm_calls = 50
llm_batch_concurrency = 8
max_output_chars = 10000
exec_timeout_ms = 180000
python_command = "python3"
verbose = true

Environment variables (CODEX_NATIVE_RLM*) are still supported and will override config values when set.

Current approvals requirement

At the moment, Native RLM workflows expect full tool/file access. In the Codex TUI, run /approvals and choose full access before testing RLM loops. If approvals are stricter, tool execution and file editing may fail mid-loop.

---

Quickstart

Installing and running Codex CLI

Install globally with your preferred package manager:

# Install using npm
npm install -g @openai/codex

# Install using Homebrew
brew install --cask codex

Then simply run codex to get started.

You can also go to the latest GitHub Release and download the appropriate binary for your platform.

Each GitHub Release contains many executables, but in practice, you likely want one of these:

• macOS
  • Apple Silicon/arm64: codex-aarch64-apple-darwin.tar.gz
  • x86_64 (older Mac hardware): codex-x86_64-apple-darwin.tar.gz
• Linux
  • x86_64: codex-x86_64-unknown-linux-musl.tar.gz
  • arm64: codex-aarch64-unknown-linux-musl.tar.gz

Each archive contains a single entry with the platform baked into the name (e.g., codex-x86_64-unknown-linux-musl), so you likely want to rename it to codex after extracting it.

Using Codex with your ChatGPT plan

Run codex and select Sign in with ChatGPT. We recommend signing into your ChatGPT account to use Codex as part of your Plus, Pro, Team, Edu, or Enterprise plan. Learn more about what's included in your ChatGPT plan.

You can also use Codex with an API key, but this requires additional setup.

Docs

• Codex Documentation
• Contributing
• Installing & building
• Open source fund

This repository is licensed under the Apache-2.0 License.