Microsoft Build 2025 is on this week and one of the big announcements from the event is about GitHub. Those who write and maintain code in repositories are now getting enhanced capabilities in the form of a significant update for Copilot.
GitHub are introducing an enterprise-ready coding agent. This new agent is designed to integrate seamlessly into GitHub and VS Code, creating a powerful loop for developers. It’s clear GitHub is serious about empowering developers with AI, while also keeping them in control.
One of the big takeaways is GitHub’s commitment to developer choice. They’re even open-sourcing Copilot Chat in VS Code (my favourite IDE) and expanding GitHub Models to support new AI models, including Grok 3 from xAI. This move is all about making the platform more versatile and accessible to a wider range of developers.
Integrated, Secure, and Fully Customisable
The new Copilot coding agent works right within GitHub’s existing control layers. You can assign it a GitHub issue or kick it off from Copilot Chat in VS Code, and it gets to work. As it progresses, it’ll push commits to a draft pull request, and you can track its progress through the agent session logs.
This agent is built with security front and centre, preserving your current security posture. It includes features like branch protections and controlled internet access to ensure development workflows are safe and compliant. Crucially, its pull requests need human approval before any Continuous Integration (CI)/Continuous Deployment/Delivery (CD) workflows run, adding an extra layer of protection.
Powered by GitHub Actions
Every good software engineering agent needs a robust compute environment, and the Copilot coding agent leverages GitHub Actions. This means it spins up a secure, fully customisable development environment, benefiting from the world’s largest CI/CD ecosystem with over 25,000 actions.
Microsoft have released a promo video for the new features, check it out below.
Enterprise-Ready at its Core
GitHub has been privately previewing this agent with internal teams and select customers, and the results are promising. It’s proving to be excellent for low-to-medium complexity tasks in well-tested codebases, whether it’s adding features, fixing bugs, or improving documentation. This means developers can offload more routine tasks and focus on higher-value work.
“We believe agentic AI is the next phase of evolution for software engineering. Together with GitHub, we have tightened DevSecOps cycles, while using existing security controls without modification.
Rodrigo Castillo, Chief Technology Officer at CommBank
We’re also automating common tasks, such as tech debt management, giving our engineering teams more time for creative flow: designing better products and experiences for our customers, and better platforms for our other engineers.”
“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.”
Alex Devkar, Senior Vice President, Engineering and Analytics at Carvana,,
“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.”
James Zabinski, DevEx Lead at EY.
The Copilot coding agent is now available in preview to all Copilot Enterprise and Copilot Pro+ users. Using the agent consumes Copilot premium requests from your entitlements, along with GitHub Actions minutes, both of which have an included allowance.
Even More AI Capabilities
GitHub is also rolling out a bunch of other AI-powered features to make development even more flexible and efficient. They’re really pushing the boundaries of what’s possible with AI in the dev world.
Opening up Copilot
Starting next month, the GitHub Copilot Chat extension will be open-sourced under an MIT license, with key AI capabilities gradually integrated into VS Code core. This means developers will soon have more control over how AI works in their editor.
Extended GitHub Models
Every GitHub user can now enable the Models tab in any repository to build, test, and manage AI features in one place. This provides developers with prompt management, lightweight evaluations, and enterprise controls, allowing them to experiment with and deploy industry-leading models on GitHub, all with built-in governance and security.
Welcoming Grok to GitHub Models
Developers now have more model choices on GitHub, with xAI’s Grok 3 now available in GitHub Models. Creator of xAI, Elon Musk, made his Microsoft Build debut (virtually) on stage at Build 2025 to discuss Grok 3 on Azure with Satya Nadella, the CEO of Microsoft.
Expanded agent mode
Agent mode is rolling out in JetBrains, Eclipse, and Xcode in public preview to all Copilot users, extending Copilot’s reach across various development environments.
For more information, head to https://github.com/
