During the Microsoft Build conference, GitHub announced they’re giving Copilot agent capabilities to automate DevOps loops.
When needed, Copilot will be able to essentially take the wheel – that’s the gist of their new asynchronous coding agent, now embedded directly within GitHub and chattable from VS Code. It’s all about creating what they’re calling a “powerful Agentic DevOps loop” for developers.
Alongside this new agent, GitHub announced they’re open-sourcing Copilot Chat in VS Code. The platform is getting a boost with fresh capabilities in GitHub Models, and that includes nodding to xAI by bringing support for Grok 3. For those not living entirely in the VS Code world, agent mode is also making its way to JetBrains, Eclipse, and Xcode.
Thomas Dohmke, CEO of GitHub, said: “GitHub is where the world’s developers work on their projects. Now, it’s becoming the place where they collaborate with agents in a configurable, steerable, and verifiable way. It’s vital that organisations and developers are ready to embrace these agents without compromising their security posture.
“Built around an integrated, secure, and fully customizable development environment powered by GitHub Actions, the Copilot coding agent is the most enterprise-ready of its kind—amplifying human developers with trust by design. And these protections aren’t just for us: as the new home of AI agents, we’re making the same primitives available to partners to ensure an open ecosystem for agentic peer programming.”
How does this new AI teammate work?
The Copilot coding agent isn’t some rogue entity; it’s designed to slot right into the existing GitHub workflow and work within its inherent control systems throughout the software development lifecycle. You can kick things off by assigning a GitHub issue directly to Copilot, or by giving it a task via Copilot Chat in VS Code.
As this AI agent gets to work, it’s not operating in a black box. It will push its commits to a draft pull request, and developers can keep an eye on its progress every step of the way through dedicated agent session logs. Humans are still very much in the loop and developers can offer feedback and ask the agent to rethink or iterate on its approach through the standard pull request review process.
Understandably, letting an AI loose on your codebase raises security questions. GitHub seems to have anticipated this. The agent is built to respect existing security setups, with safety nets like branch protections and controlled internet access baked in to keep development workflows secure and above board. What’s more, any pull requests the agent cooks up still need that vital human thumbs-up before any CI/CD magic happens, offering an extra checkpoint for the build and deployment pipeline.
For teams wanting to give their agent a bit more worldly knowledge, the Model Context Protocol (MCP) allows them to grant the coding agent access to data and tools from outside the GitHub ecosystem. These MCP servers can be set up in the repository’s settings, making the agent even more powerful.
Kate Holterhoff, Senior Analyst at RedMonk, commented: “With its autonomous coding agent, GitHub is looking to shift Copilot from an in-editor assistant to a genuine collaborator in the development process. This evolution aims to enable teams to delegate implementation tasks and thereby achieve a more efficient allocation of developer resources across the software lifecycle.”
GitHub powers Copilot agent capabilities with Actions
For any AI software engineering agent to do its job, it needs a digital workshop. The Copilot coding agent sets up shop using a secure and customisable development environment powered by the familiar GitHub Actions.
If you’ve been around the GitHub block, you’ll know Actions. Launched back in 2018, it’s become the largest CI/CD ecosystem out there with 25,000+ actions available in the GitHub Marketplace. On any given weekday, those GitHub-hosted and self-hosted runners are chugging through over 40 million jobs.
By tapping into Actions, Copilot is using a compute platform that’s already proven its salt in terms of reliability and security, trusted by everyone from open-source heroes to large enterprises.
GitHub has been putting this agent through its paces internally and with some select customers. The early verdict? It shines when it comes to those low-to-medium complexity tasks, especially in codebases that are already well-tested.
Alex Devkar, Senior Vice President of Engineering and Analytics at Carvana, said: “The GitHub Copilot coding agent fits into our existing workflow and converts specifications to production code in minutes. This increases our velocity and enables our team to channel their energy toward higher-level creative work.”
James Zabinski, DevEx Lead at EY, added: “The Copilot coding agent is opening up doors for human developers to have their own agent-driven team, all working in parallel to amplify their work. We’re now able to assign tasks that would typically detract from deeper, more complex work—allowing developers to focus on high-value coding tasks.”
Starting today, if you’re a Copilot Enterprise or Copilot Pro+ user, you can get your hands on this agent in preview. Using it will dip into your Copilot premium requests and also use up some GitHub Actions minutes.
See also: Rust developers offered $20k for rav1d to reach C performance
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.
Explore other upcoming enterprise technology events and webinars powered by TechForge here.
