Skip to main content

​
OpenAI Codex CLI Integration

AiHubMix offers seamless integration with OpenAI Codex CLI, enabling you to leverage advanced AI programming assistance directly in your command line environment. With simple configuration steps, you can use natural language to perform various programming and system operations from your terminal.
Before using, please run the following command to install or update: 
npm install -g @openai/codex

​
Configuration Steps

​
1. Set Environment Variables

Open your shell configuration file (e.g., .zshrc or .bashrc) and add the following environment variables: 
export OPENAI_BASE_URL="https://aihubmix.com/v1"
export OPENAI_API_KEY="sk-***"
The OPENAI_API_KEY should be your AiHubMix key. The variable name remains OPENAI_API_KEY for compatibility with the OpenAI native client.
Skip manual environment variable configuration – use the graphical tool for quick setup: Configuration via CC-Switch.

​
2. Apply Configuration Changes

Execute the following command in your terminal to apply the environment variables: 
source ~/.zshrc  # If using zsh
# or
source ~/.bashrc  # If using bash

​
3. Launch Codex CLI

Navigate to your project directory and run the codex command: 
cd /your_project_path
codex
Image

​
4. Execute Tasks Using Natural Language

Now you can input instructions to Codex CLI using natural language, for example: 
# Sample input
Break down the TTSComponent, and distill its core logic into a Mermaid flowchart.
Image

​
Configuration via cc-switch

  1. Run CC-Switch and click “Add Provider.”
Image
  1. From the preset list, select “AiHubMix.”
Image
  1. Enter your API key in the “API Key” field, then click “Add” to save the settings.
Image
  1. Return to the home page, select “AiHubMix” from the vendor list, and click “Start” to begin using it.
Image

​
Advanced Configuration

  • The default model is codex-mini-latest - a fine-tuned version of o4-mini specifically for use in Codex CLI, which can be modified in ~/.codex/config.json
  • Currently only OpenAI models are supported, you can find the model list in the Responses API documentation
  • You can customize the system prompt by editing the ~/.codex/instructions.md file to tailor the AI assistant’s behavior

​
Useful Command Reference

​
Help Command

codex -h

​
Complete Command Options

  Usage
    $ codex [options] <prompt>

  Options
    -h, --help                 Show usage and exit
    -m, --model <model>        Model to use for completions (default: codex-mini-latest)
    -i, --image <path>         Path(s) to image files to include as input
    -v, --view <rollout>       Inspect a previously saved rollout instead of starting a session
    -q, --quiet                Non-interactive mode that only prints the assistant's final output
    -a, --approval-mode <mode> Override the approval policy: 'suggest', 'auto-edit', or 'full-auto'

    --auto-edit                Automatically approve file edits; still prompt for commands
    --full-auto                Automatically approve edits and commands when executed in the sandbox

    --no-project-doc           Do not automatically include the repository's 'codex.md'
    --project-doc <file>       Include an additional markdown file at <file> as context
    --full-stdout              Do not truncate stdout/stderr from command outputs

  Dangerous options
    --dangerously-auto-approve-everything
                               Skip all confirmation prompts and execute commands without
                               sandboxing. Intended solely for ephemeral local testing.

  Experimental options
    -f, --full-context         Launch in "full-context" mode which loads the entire repository
                               into context and applies a batch of edits in one go. Incompatible
                               with all other flags, except for --model.

  Examples
    $ codex "Write and run a python program that prints ASCII art"
    $ codex -q "fix build issues"