OpenAIfeature
September 29, 2026·Source

Codex Security Cloud (Research Preview)

A Codex plugin that scans your GitHub repositories for security vulnerabilities in the cloud, confirms them, and drafts fixes, for Pro, Business, Enterprise, and Edu users.

Codex Security Cloud (Research Preview) announcement image
  • Runs a one-time repository scan or watches new commits, with an optional look back over recent history; scanning can be paused
  • Builds an editable threat model, scans, then tries to reproduce each issue in an isolated container to cut false positives
  • Results are ranked by severity with evidence and fix guidance; Fix with Codex opens a draft pull request, and patches are never applied automatically
  • OpenAI positions it as a complement to existing code-scanning tools, not a replacement
  • Includes access to OpenAI's Daybreak Blue defensive-security models without a separate application
  • Research preview on web and desktop for Pro, Business, Enterprise, and Edu

Best for

Finding and confirming real vulnerabilities across a codebase without a dedicated security engineer.

When to use it

You want every new commit checked, or a one-time audit of a repo before launch.

Who benefits

Small engineering teams and security leads at companies shipping their own software.

Join the newsletter to get ahead with AI.

I share strategies, new modules, free guides, and tools to get you better at using AI in your work.

Leading a team? See team programs →