GitHub previews support for Claude and Codex coding agents

GitHub is adding support for the Anthropic Claude and OpenAI Codex coding agents, via its Agent HQ AI platform. The capability is in public preview.

Copilot Pro+ and Copilot Enterprise users now can run multiple coding agents directly inside GitHub, GitHub Mobile, and Visual Studio Code, GitHub announced on February 4. GitHub said that Copilot CLI support was coming soon.

With Claude, Codex, and GitHub Copilot in Agent HQ, developers can move from idea to implementation using different agents for different steps without switching tools or losing context, the company said. “We’re bringing Claude into GitHub to meet developers where they are,” said Katelyn Lesse, head of platform at Anthropic, in a statement. “With Agent HQ, Claude can commit code and comment on pull requests, enabling teams to iterate and ship faster and with more confidence. Our goal is to give developers the reasoning power they need, right where they need it.”

Working directly with these agents in GitHub and VS Code, users can do the following, according to GitHub: 

• Explore trade-offs early by running agents in parallel to surface competing approaches and edge cases before code hardens. 
• Keep context attached to the work, with agents operating inside the user’s repository, issues, and pull requests instead of starting from stateless prompts. 
• Avoid new review processes, with agent-generated changes showing up as draft pull requests and comments, reviewed the same way a user would review a teammate’s work. 

GitHub’s Agent HQ also lets users compare how different agents approach the same problem, GitHub said. Developers can assign multiple agents to a task, and see how Copilot, Claude, and Codex reason about trade-offs and arrive at different solutions.