GitHub Launches Agent HQ to Unify Multiple AI Coding Agents in One Dashboard

GitHub is introducing a new hub called Agent HQ, giving developers access to multiple third-party AI coding agents directly within the platform. In addition to GitHub’s own AI coding agent, users will soon be able to experiment with models from OpenAI’s Codex, Anthropic’s Claude, Google’s Jules, xAI, and Cognition’s Devin. Developers with a GitHub Copilot subscription will gain access to a new “mission control” dashboard, designed to help them manage, monitor, and control multiple AI agents simultaneously. According to GitHub COO Kyle Daigle, this dashboard serves as a centralized control plane for all AI agent activity on the platform, whether it’s the native GitHub agent or one from a partner. One of the key features of Agent HQ is the ability to run several AI agents in parallel on the same task. This allows developers to compare outputs and select the best solution, improving both efficiency and code quality. Before the official launch, OpenAI’s Codex is already available to Copilot Pro Plus users who are part of the VS Code Insiders program. Alongside Agent HQ, GitHub is rolling out several new tools to enhance the AI coding experience. A new Plan Mode in VS Code leverages Copilot to generate a detailed, step-by-step roadmap for a given development task, which the AI agent can then execute. Additionally, GitHub is introducing a built-in code review step within Copilot, enabling the AI to use advanced tools like CodeQL to analyze code for security flaws, performance issues, and best practices before presenting it to the developer. These updates reflect GitHub’s broader strategy to evolve from a code hosting platform into a full-stack AI-powered development environment, empowering developers with smarter, more collaborative tools.