Recently, the AI developer community has seen a major update. OpenAI officially released an open-source project called codex-plugin-cc on GitHub. This plugin allows developers to directly call the Codex model capabilities of OpenAI within Anthropic's command-line development tool, Claude Code.

This "cross-brand" integration breaks down the previous ecological barriers between large model tools, allowing developers to simultaneously utilize the technical advantages of two industry giants without switching environments. With simple instruction configurations, Claude Code instantly becomes a versatile programming assistant that combines the strengths of both companies.

image.png

Empowered by this plugin, users can initiate a standard read-only code review using the /codex:review command and receive professional improvement suggestions from Codex. This dual verification mechanism effectively captures logic vulnerabilities that a single model might miss, adding a "double insurance" for code quality.

More notably, it features an "adversarial review" function, allowing developers to use /codex:adversarial-review to actively request Codex to challenge existing design decisions. This mode is specifically used to stress-test the rationality of system architecture, uncovering potential performance bottlenecks or security risks from a "critique" perspective.

The plugin also introduces a task delegation mechanism, allowing users to assign complex debugging or repair tasks to a Codex sub-agent using /codex:rescue. This collaborative mode enables automatic task distribution, letting the main model and auxiliary model each play their respective roles in their areas of expertise.

github:https://github.com/openai/codex-plugin-cc