OpenAI launches Codex Security AI Agent to detect and fix complex code vulnerabilities

OpenAI has introduced a new application security agent called Codex Security, designed to analyze software projects in depth, detect complex vulnerabilities, and provide recommended fixes.

As reported by AzerNEWS, according to the company, the new tool is being released in research preview and is available starting today to customers of ChatGPT Enterprise, Business, and Edu through the Codex web platform. OpenAI noted that the feature will be free to use for the next month.

In a statement, the company said Codex Security combines advanced agentic reasoning from its frontier AI models with automated validation to deliver high-confidence vulnerability findings and practical remediation steps.

“By combining agentic reasoning from our frontier models with automated validation, it delivers high-confidence findings and actionable fixes so teams can focus on the vulnerabilities that matter and ship secure code faster,” the company said.

The tool was initially developed last year under a private beta program called Aardvark. It is designed to build full system context, generate editable threat models, prioritize and verify security issues, and help developers fix vulnerabilities using comprehensive system analysis.