Codex CLI Setup Guide

1. Open the config directory

Codex's config directory is .codex, located in your home directory. Open it according to your OS:

Windows

Press Win + R, enter the following and hit Enter:

%userprofile%\.codex

macOS

# Option 1: open directly from the terminal
open ~/.codex

# Option 2: in Finder press Cmd + Shift + G, enter ~/.codex and hit Enter

Linux

# Enter the config directory in a terminal
cd ~/.codex

# To open in a file manager (GNOME example)
xdg-open ~/.codex

You'll see the directory contents. If there's no config.toml / auth.json, create them manually and open them.

TIP

If the .codex directory doesn't exist yet, run codex once in the terminal to auto-generate it, or create the directory manually.

2. Edit config.toml

config.toml is the core config file — used to configure the API URL, the model, and features like reasoning depth.

Write the following into config.toml:

model_provider = "cctq_codex"
model = "gpt-5.6-sol"
plan_mode_reasoning_effort = "xhigh"
model_reasoning_effort = "high"
disable_response_storage = true
supports_websockets = false

[model_providers.cctq_codex]
name = "cctq_codex"
base_url = "https://www.cctq.ai/v1"
wire_api = "responses"
requires_openai_auth = true

• base_url: the relay request URL — set it to https://www.cctq.ai/v1 (must end with /v1).
• model: the model to use — set as needed.

3. Edit auth.json

auth.json holds your API key. Write the following into auth.json:

{
  "OPENAI_API_KEY": "sk-***"
}

• OPENAI_API_KEY: your token — replace sk-*** with your actual key.

Save, reopen the terminal, and run codex to start.