A Dockerised environment and Makefile for interacting with the OpenAI Codex (via the OpenAI CLI) inside a container. This setup isolates your local development environment from the Node.js and OpenAI CLI dependencies.

• Dockerfile: Defines a Node.js container with the OpenAI CLI installed.
• docker-compose.yml: Runs the service with environment variables loaded from a given .env file.
• Makefile: Provides shortcuts for building, running, and interacting with Codex via Docker:
  • Build and manage containers.
  • Shell access.
  • Dependency install & update.
  • Codex completions and edits.
• .env.example: Template for environment variables.

• Docker (v20+)
• Docker Compose plugin (use docker compose)
• An OpenAI API key

1. Clone the repository

   git clone https://github.com/gocanto/openai-codex.git
   cd openai-codex

2. Copy and configure .env

   cp .env.example .env

   • OPENAI_API_KEY: Your OpenAI secret key
   • NPM_TOKEN: (Optional) Your npm access token for private packages

# either export it directly
export OPENAI_API_KEY="your-api-key-here"

# or put it in a .env file alongside your docker-compose/Makefile:
OPENAI_API_KEY=your-api-key-here

make build   # Build the Docker image
make up      # Start containers in detached mode

• make help : See all the available commands
• make build : Build the Docker image
• make up : Start containers in detached mode
• make logs : View container logs
• make shell : Open an interactive shell in the container
• make install: Install npm dependencies (npm ci) inside container
• make update : Update npm dependencies (npm update) inside container
• make start : Run your application (npm start)
• make down : Stop and remove containers

• Complete from prompt

  make codex-complete prompt="<Describe your coding task>"

• Edit existing file

  make codex-edit file=path/to/file prompt="<Edit instructions>"

Outputs (edits or completions) appear in your terminal or overwrite the target file (with a .edited suffix if configured).

Feel free to submit issues or pull requests for enhancements or bug fixes.

This project is licensed under the MIT Licence. See LICENSE for details.