GitHub Launches Agent HQ to Bring Multiple AI Coding Agents to Developers
Key Points
- GitHub introduces Agent HQ, a hub for multiple AI coding agents.
- Developers can access OpenAI Codex, Anthropic Claude, Google Jules, xAI, and Cognition Devin.
- A mission‑control dashboard lets users manage and track AI agents.
- Parallel execution allows side‑by‑side comparison of agent outputs.
- VS Code gains Plan Mode for task planning and a code‑review step with CodeQL.
- Early access to Codex is provided to Copilot Pro Plus Insiders users.
GitHub is rolling out a new platform called Agent HQ that lets developers use a range of AI coding agents—including OpenAI’s Codex, Anthropic’s Claude, Google’s Jules, xAI, and Cognition’s Devin—directly within GitHub. Available to Copilot subscribers, the service adds a mission‑control style dashboard, the ability to run several agents in parallel on the same task, and new VS Code tools such as Plan Mode and a code‑review step powered by CodeQL. The move expands the ecosystem of AI‑assisted development while giving users more control over which agent’s output they prefer.
New Agent HQ Platform
GitHub announced the upcoming launch of Agent HQ, a centralized hub that expands the AI coding experience beyond the native GitHub Copilot assistant. The platform will allow developers to access third‑party AI coding agents from several leading providers without leaving the GitHub environment.
Expanded AI Agent Options
In addition to Copilot, developers will soon be able to try OpenAI’s Codex, Anthropic’s Claude, Google’s Jules, xAI’s offering, and Cognition’s Devin. These agents will become available to users who have a GitHub Copilot subscription, giving them a broader palette of AI tools for code generation and assistance.
Mission Control Dashboard
GitHub is introducing a “mission control” dashboard that serves as a control plane for managing multiple AI agents. The dashboard lets users select, configure, and track the performance of each agent, providing a unified interface for AI‑driven development workflows.
Parallel Agent Execution
Agent HQ will enable developers to run several AI agents in parallel on the same coding task. This feature lets users compare outputs side‑by‑side and choose the solution that best fits their needs, effectively turning the platform into a testing ground for different AI approaches.
Enhanced VS Code Features
GitHub is also adding new capabilities to Visual Studio Code. A “Plan Mode” will use Copilot to generate a step‑by‑step plan for a task, which the AI can then execute. Additionally, a code‑review step will integrate tools like CodeQL, allowing the AI to evaluate generated code for security and quality before presenting it to the developer.
Early Access and Availability
Ahead of the full Agent HQ rollout, OpenAI’s Codex is being made available to Copilot Pro Plus users enrolled in the VS Code Insiders program. The broader suite of agents and dashboard features will follow in the coming months, expanding the AI toolkit available to GitHub’s developer community.