Security teams on GitHub with ChatGPT Pro or business plans can try it in research preview, but token rates aren't published.

Kenning's take

What it does

Codex Security connects to GitHub repositories, builds a threat model of the code and scans the full repository or new commits for vulnerabilities. It tries to confirm each finding in an isolated environment and suggests patches for the team to review. It is a research preview for ChatGPT Pro, Business, Enterprise and Edu users.

• Editable threat model per repository
• Full-repository and commit scans
• Validates findings in an isolated environment
• Proposed patches for review
• Role-based access controls for Enterprise and Edu

Who it suits. Security and engineering teams who want automated vulnerability review of their code.

Key facts

Category

Developer tools

Part of

ChatGPT

Made by

OpenAI

Released

6 March 2026

How to get it

Research preview for ChatGPT Pro, Business, Enterprise and Edu users through Codex on the web; GitHub repositories only

Works on

Web

API price

Token-based billing for Codex Security Cloud; rates not published on the help page

Pricing

Plan	Price	Billed
ChatGPT Pro	$100	month (lowest Pro tier)
ChatGPT Business	$20	user per month, billed annually

source

FAQ

Facts from the maker's own pages and the sources listed. Checked 2026-10-11.