OpenAI introduced Codex Security, a command‑line interface and TypeScript SDK designed to detect, validate, and remediate security vulnerabilities in source code. The tool requires Node.js 22 or later, Python 3.10 or later, and access to Codex Security. Users can install it via npm and authenticate with either a ChatGPT sign‑in or an API key. CI pipelines can use an API key directly. Scan commands, such as `npx codex-security scan .`, can specify authentication method with `--auth chatgpt` or `--auth api-key`. If both credentials are present, interactive scans prompt the user to choose; noninteractive scans prioritize the API key. To set the ChatGPT credential as default, users must unset configured API keys. Scan state is stored in a designated workbench directory; if write access is unavailable, the `CODEX_SECURITY_STATE_DIR` environment variable can be set to an alternative path. The TypeScript SDK allows developers to integrate scanning into applications by importing the `CodexSecurity` class and invoking its `run` method. OpenAI provides comprehensive documentation covering installation, authentication, scan options, and continuous integration setup.