Introducing Claude and Codex: GitHub’s AI Coding Agents

GitHub has announced the integration of Claude by Anthropic and OpenAI’s Codex AI coding agents directly into GitHub. This exciting development, released today, brings Claude and Codex to GitHub, GitHub Mobile, and Visual Studio Code for users subscribed to Copilot Pro Plus or Copilot Enterprise.

This move is part of Agent HQ, GitHub’s initiative to seamlessly integrate AI agents into developers’ daily workflows on GitHub. With the addition of Copilot, Claude, Codex, and other custom agents, developers now have the flexibility to choose their preferred coding assistant when working on tasks. Each coding agent will utilize a premium request, allowing developers to assign agents to specific issues and pull requests.

Developers will have the ability to evaluate the performance of Copilot, Claude, and Codex, and assess how each AI coding agent generates solutions. GitHub’s Chief Product Officer, Mario Rodriguez, emphasizes the importance of reducing context switching in software development. With Codex, Claude, and Copilot available in Agent HQ, developers can seamlessly transition from idea to implementation using different agents for various tasks without disrupting their workflow.

GitHub continues to embrace rival AI models and agents to enhance its Copilot offering. Developers can already access models from Anthropic, Google, xAI, and OpenAI in GitHub Copilot, making the integration of rival AI coding agents a logical progression.

Access to Claude and Codex will soon be available to more GitHub Copilot subscription types. GitHub is also collaborating with Google, Cognition, and xAI to introduce additional agents to GitHub, Visual Studio Code, and the Copilot CLI.

As Microsoft explores Anthropic’s Claude Code tool, developers within the company are comparing Claude Code with GitHub Copilot to enhance the latter’s capabilities. This collaboration underscores the industry’s commitment to advancing AI technologies for software development.