Added Codex and Claude coding agents to GitHub

GitHub has announced that starting today it will make coding artificial intelligence agents Claude from Anthropic and Codex from OpenAI available to developers directly within its platform.

A new public preview for users with a Copilot Pro Plus or Copilot Enterprise subscription adds Claude and Codex agents to GitHub, GitHub Mobile, and Visual Studio Code, The Verge reports.

More coding agents are added to GitHub

This move is part of Agent HQ, or GitHub’s vision to integrate artificial intelligence agents natively into the daily workflow of developers. Developers can now choose between Claude, Codex, and Copilot or other custom agents when creating a task.

Each Coding Agent consumes a Premium Request, and developers can assign agents to Issue and Pull Requests.

Developers can also evaluate the performance of Copilot, Claude, and Codex and compare how well each AI agent has produced a solution. “With Codex, Claude and Copilot in Agent HQ, you can go from idea to implementation using different agents for different stages,” said Mario Rodriguez, GitHub’s senior product manager. Without changing tools or losing work context.

GitHub has been quick to welcome competing AI models and agents to improve its Copilot service. Developers can now access models from Entropic, Google, OpenAI, and xAI on GitHub Copilot.

Access to Claude and Codex will soon be extended to more types of GitHub Copilot subscriptions. GitHub is also working with Google, xAI, and Cognition to add more agents to Visual Studio Code and the Copilot CLI.

GitHub’s integration of competing AI coding agents coincided with Microsoft’s increased testing of Entropy’s Claude Code tool. For this purpose, Microsoft developers have been asked to compare Claude Code with GitHub Copilot, which is an attempt to improve GitHub Copilot.