Skip to content
Unlock AI’s true impact across the SDLC. Explore key findings from Gartner®.

February ‘26 enterprise roundup

In case you missed it…

Published via GitHub Executive Insights | Authored by Dave Burnison

Welcome to the February, 2026 edition of the GitHub Monthly Enterprise Roundup (MER). Key updates in this edition include a collection of GitHub Platform improvements aimed at making it easier to manage GitHub Enterprise at scale, the launch of Agent HQ in public preview, the launch of the GitHub Copilot Software Development Kit (SDK) and significant updates to the GitHub Copilot Command Line Interface (CLI). 

The audience for the MER is anyone in enterprise software development so, there is a wide range of information here. We don't expect you to read every word. Skim through the sections that apply to how you use GitHub and dig into links that are the most relevant to you. Since some readers may skip over entire sections, you may see the same link appear in multiple sections such as a link that applies to both Code Security and CI/CD. Any one person will not read every link in this post but, across your team, every link may be read by at least one of your team members. Pass this MER along to your colleagues or pass along specific links that will be beneficial to others. 

Want to get notified of when the next MER is available? Go to GitHub Enterprise on LinkedIn and click on the "Follow" button. In addition to MER notifications you'll be notified when new episodes of GitHub at Work Podcast and other enterprise focused content becomes available. 

GitHub Platform

We have been listening to our enterprise customers for years. We are excited to share product updates and new guidance to assist those who manage GitHub for hundreds if not thousands of stakeholders. This month's updates demonstrate how we are acting on your feedback to address the issues in multiple areas you face managing GitHub Enterprise at scale. 

Developer Skills

General developer expertise based on our own experience and the collective experience of our customers and partners. It's time to start diving into how AI is going to work alongside you to make you a better, more productive developer, not replace you. Check out the new posts 📢, documentation 📄, and videos 📺 to see how AI can make you an awesome developer and guidance for how large enterprises should approach adopting AI. Also, learn from the wisdom of Anders Hejlsberg whose work has shaped how millions of developers code. 

  • 📢 How to maximize GitHub Copilot's agentic capabilities - GitHub Copilot’s agentic capabilities can fundamentally change how your teams approach architecture, refactoring, migrations, and multi-file coordination—turning Copilot into a true systems‑level collaborator rather than just an autocomplete tool. This guide shows how senior‑engineer workflows like modularization, safe schema evolution, and cross‑module refactoring can be accelerated and de‑risked with agent mode. If you lead complex systems or steward legacy codebases, you need to know how these capabilities can multiply engineering impact while strengthening technical rigor.

  • 📄 Using GitHub Copilot coding agent to improve a project - This guide shows how Copilot’s coding agent can actively reduce technical debt, streamline environment setup, and automate issue creation—freeing your team to focus on higher‑value engineering work. It explains how to turn neglected or aging codebases into maintainable, well‑structured projects by letting Copilot surface problems and generate targeted fixes. For teams under pressure to do more with less, it demonstrates a repeatable way to accelerate improvement without slowing ongoing delivery.

  • 📢 & 📺 What AI is actually good for, according to developers - Developers are increasingly clear about what they need from AI: tools that preserve flow, reduce friction, and stay out of the way while helping them focus on meaningful work. This post explains why AI that integrates into existing workflows—not flashy chat interfaces—actually boosts productivity, trust, and control for engineering teams. Understanding these insights helps enterprise developers and leaders evaluate which AI capabilities genuinely improve delivery speed and developer experience versus those that add noise.

  • 📢 & 📺 7 learnings from Anders Hejlsberg: The architect behind C# and TypeScript (35:14) - Anders Hejlsberg distills four decades of language and tooling design into lessons that directly impact how modern engineering teams build at scale—fast feedback loops, behavioral compatibility, and pragmatic evolution over purity. For enterprise developers, his insights explain why certain tools endure under real-world pressure and how AI‑driven workflows make semantic accuracy more critical than ever. If you’re responsible for choosing, scaling, or future‑proofing your organization’s developer stack, this piece highlights the architectural patterns you need to understand to avoid costly missteps.

    • Being in the enterprise space, I really appreciate these comments, "Any system that needs to scale across teams requires a shift from personal taste to shared outcomes. The goal stops being code that looks the way you would write it, and starts being code that many people can understand, maintain, and evolve together. … Languages do not succeed because they are perfectly designed. They succeed because they accommodate the way teams actually work." This mindset goes well beyond programming languages and applies to everything that is in your enterprise SDLC and DevOps tool stack. 

AI & ML - GitHub Copilot

Recent advancements and feature updates for GitHub Copilot, with a particular focus on coding agent choice, the CLI and the GitHub Copilot SDK. Checkout the guidance in the blog posts 📢 and new documentation 📄 on how to get the kind of output from GitHub Copilot you want and need to drive real business value.

GitHub Copilot coding agent

  • 📢 & 📺 Pick your agent: Use Claude and Codex on Agent HQ (1:21) - Agent HQ now lets your teams run GitHub Copilot, Claude, and Codex side‑by‑side—directly inside GitHub and VS Code—so you can compare reasoning, automate reviews, and move from idea to implementation without losing context. AI-assisted development is shifting from individual productivity to organizational scale, giving you governance, code-quality checks, auditability, and impact metrics baked into your existing GitHub workflows. If you’re responsible for velocity, quality, or architectural rigor, this update shows how multi‑agent workflows can transform the way your org builds and reviews software. NOTE: No additional subscriptions are required. Access to Claude and Codex is included with your existing Copilot Enterprise subscription. Each coding agent session consumes one premium request during public preview. See also 🚢 Claude and Codex are now available in public preview on GitHub

  • 📄 Using GitHub Copilot coding agent to improve a project - This guide shows how Copilot’s coding agent can actively reduce technical debt, streamline environment setup, and automate issue creation—freeing your team to focus on higher‑value engineering work. It explains how to turn neglected or aging codebases into maintainable, well‑structured projects by letting Copilot surface problems and generate targeted fixes. For teams under pressure to do more with less, it demonstrates a repeatable way to accelerate improvement without slowing ongoing delivery.

  • 📢 How to maximize GitHub Copilot's agentic capabilities - GitHub Copilot’s agentic capabilities can fundamentally change how your teams approach architecture, refactoring, migrations, and multi-file coordination—turning Copilot into a true systems‑level collaborator rather than an autocomplete tool. This guide shows how senior‑engineer workflows like modularization, safe schema evolution, and cross‑module refactoring can be accelerated and de‑risked with agent mode. If you lead complex systems or steward legacy codebases, you need to know how these capabilities can multiply engineering impact while strengthening technical rigor.

  • 🚢 Introducing the Agents tab in your repository - A streamlined, in‑repository mission‑control experience now brings all your Copilot agent sessions, logs, and task workflows together in one intuitive place to help your team move faster and stay focused.

  • 📢 Building an agentic memory system for GitHub Copilot - GitHub is introducing a cross‑agent memory system that lets Copilot learn from your codebase and development patterns—so every agent becomes more accurate, consistent, and context‑aware over time. This matters because it eliminates repetitive re‑explanation across tools and teams, enabling safer automation, higher‑quality code reviews, and measurable productivity gains. If you want to understand how AI agents will soon retain, verify, and share knowledge across your entire SDLC, this is the post to read. It’s important to note that memories are tightly scoped. Memories for a given repository can only be created in response to actions taken within that repository by contributors with write permissions, and can only be used in tasks on that same repository initiated by users with read permissions. Much like the source code itself, memories about a repository stay within that repository, ensuring privacy and security. See also, 🚢 Agentic memory for GitHub Copilot is in public preview.

  • 📢 & 📺 Context windows, Plan agent, and TDD: What I learned building a countdown app with GitHub Copilot (1:33:39) - This post and video show how real-world AI-assisted development goes far beyond prompting—highlighting why context management, clarifying agents, and test‑driven development are becoming essential engineering skills. By walking through a full end‑to‑end build, it demonstrates how Copilot’s Plan and custom agents help teams uncover hidden requirements, reduce rework, and catch subtle bugs long before they reach users. If you’re guiding modern development practices or scaling engineering productivity, this is a practical look at the AI‑augmented workflows your teams will need next.

GitHub Copilot CLI

  • 📢 & 📺 Power agentic workflows in your terminal with GitHub Copilot CLI (14:39) - See how the GitHub Copilot CLI turns your terminal into a fully agent‑powered development environment, enabling you to plan, implement, and automate complex workflows without leaving your command line. For enterprise developers and engineering leaders, it explains why agentic workflows aren't just a productivity boost—they’re a strategic shift that reduces context switching, accelerates delivery, and brings AI‑driven consistency and governance directly into existing DevOps processes. In short, if you want to understand how AI is reshaping secure, scalable software delivery from the terminal up, this is essential reading.

  • 🚢 GitHub Copilot CLI: Enhanced agents, context management, and new ways to install - This update shows how the Copilot CLI is becoming a far more powerful automation and productivity engine—introducing built‑in task‑focused agents, streamlined model and context management, and flexible installation and scripting options that can immediately accelerate how your teams build and ship software.

  • 📢 A cheat sheet to slash commands in GitHub Copilot CLI - GitHub Copilot CLI’s new slash commands give developers fast, predictable, and auditable ways to run tests, fix code, manage context, and automate workflows directly from the terminal—no prompt‑crafting or tool‑switching required. For teams operating at scale, these commands bring clarity, security, and repeatability to everyday tasks, making Copilot’s behavior explicit and traceable. If you care about developer velocity, operational consistency, or safeguarding sensitive environments, this cheat sheet shows why slash commands are becoming essential to modern enterprise engineering.

  • 🚢 GitHub Copilot CLI: Plan before you build, steer as you go - GitHub Copilot CLI now gives developers far greater control and clarity in the terminal through collaborative planning, advanced reasoning options, real‑time steering, and deeper integration with existing workflows.

  • 📺 How to use the /share command in GitHub Copilot CLI | Demo (3:34) - Did you know you can share your GitHub Copilot CLI sessions with your team? In this demo, Scott Hanselman (@shanselman) explores the /share command, which exports your conversation logs and diagrams directly to a gist. We also cover model selection (Claude Opus, Sonnet, etc.) and how to visualize Next.js routing structures in the terminal. 

  • 📺 Demo: Using /delegate in the GitHub Copilot CLI (2:17) - Learn how to offload heavy lifting from your local machine to the cloud using the /delegate command in GitHub Copilot CLI. In this demo, @shanselman shows how to task an agent with upgrading a Next.js application while he continues to work. See how the agent creates a pull request, runs tests, and streams logs back to you.

  • 📺 Switching models in GitHub Copilot CLI | demo (2:53) - @shanselman demonstrates the power of choice in the GitHub Copilot CLI. In this video, see how to switch between 14 different models, including Claude Opus 4.5 and GPT 5.2 Codex, to plan and execute a complex upgrade for a Next.js application. He also covers how to adjust reasoning effort and use voice dictation for faster prompting.

  • 📺 Configuring model context protocol in the GitHub Copilot CLI | demo (4:07) - The Model Context Protocol (MCP) is like a USB port for your AI, allowing you to plug in external tools and documentation. In this video, @shanselman demonstrates how to configure an MCP server in the GitHub Copilot CLI. He uses Context 7 to pull in up-to-date Next.js documentation, enabling Copilot to generate an accurate architecture diagram. 

  • 📺 Demo: Using GitHub Copilot CLI and yolo mode (2:20) - Tired of approving every single command? @shanselman demonstrates the --yolo command in GitHub Copilot CLI, which allows you to bypass manual approvals for tool calls and path access. Learn how this mode lets the agent run autonomously and why you might want to use it in a containerized environment.

  • 🚢 ACP support in Copilot CLI is now in public preview - By adopting Agent Client Protocol (ACP), an open, extensible protocol that lets any IDE, automation system, or custom developer tool directly interact with Copilot’s agentic capabilities, this update unlocks new ways for teams to integrate AI-driven workflows at scale.

GitHub Copilot SDK

  • 📢 Build an agent into any app with the GitHub Copilot SDK - The GitHub Copilot SDK lets you embed the same production-tested agentic engine behind Copilot CLI directly into your own applications—so you can add intelligent planning, tool use, and multi‑step execution without building that infrastructure yourself. For leaders, this means a faster, safer path to creating internal agents, automating complex workflows, and unlocking AI‑driven productivity across your software organization using your existing languages and systems. If you want to stay competitive as AI-native development becomes the norm, this is a foundational capability you can start adopting today. See also, 🚢 Copilot SDK in technical preview.

  • 📺 GitHub Copilot SDK demo: Creating "Flight School" (8:51) - Chris Reddington demonstrates "Flight School," a custom Next.js application built to personalize his learning journey using the GitHub Copilot SDK. See how he leverages agentic workflows to generate daily coding challenges based on his GitHub profile, evaluate solutions against test cases, and automatically export projects to new repositories. 

IDE Related GitHub Copilot Updates

  • 📄 Copilot feature matrix - Use the Copilot feature matrix to see which GitHub Copilot features are available in these supported IDEs: Visual Studio Code, JetBrains IDEs, Visual Studio, Eclipse, Vim/Neovim and Xcode. NOTE: The GitHub Copilot feature matrix is currently in public preview and is subject to change.

  • 🚢 GitHub Copilot in Visual Studio Code v1.109 - January Release - This release brings significant improvements to GitHub Copilot in Visual Studio Code with agent-driven workflows, improvements to agent session management, and the introduction of agent support for Claude by Anthropic.

  • 🚢 GitHub Copilot in Visual Studio — January update - Copilot’s latest Visual Studio enhancements deliver more intuitive code intelligence, finer‑grained control over completions, and productivity‑boosting editor improvements that can immediately streamline enterprise‑scale development workflows.

  • 🚢 GPT-5.2-Codex is now available in Visual Studio, JetBrains IDEs, Xcode, and Eclipse - You’ll now be able to access the model in GitHub Copilot Chat on github.com, GitHub Mobile, Visual Studio Code, Visual Studio, JetBrains IDEs, Xcode, and Eclipse through the chat model picker (agent, ask, and edit modes). 

GitHub Copilot - New Models

Additional GitHub Copilot Updates

  • 📄 Support for different types of custom instructions - GitHub Copilot’s custom instructions let you shape how AI assistants behave across your repositories, teams, and development environments, ensuring consistency and control at scale. Understanding which instruction types each IDE or Copilot feature supports helps you enforce standards, improve code quality, and guide AI-generated output in ways that match your team’s engineering practices. This page clarifies those differences so you can apply the right instruction strategy for your organization’s workflows.

  • 🚢 Showing tool calls and other improvements to Copilot chat on the web - GitHub is rolling out enhancements that give teams deeper visibility, smoother workflows, and more control when collaborating with Copilot Chat directly on the web.

  • 🚢 GitHub MCP Server: New Projects tools, OAuth scope filtering, and new features - Empower your teams with dramatically streamlined project automation, smarter permission‑aware tooling, and new enterprise‑ready deployment options designed to boost velocity and reduce operational overhead across your entire software organization.

  • 📢 Continuous AI in practice: What developers can automate today with agentic CI - Continuous AI introduces a new class of automation that tackles the judgment-heavy, context-dependent engineering work traditional CI can’t touch. It shows how lightweight, natural‑language‑driven agents can continuously handle tasks like documentation drift, dependency changes, performance regressions, and semantic bugs—freeing developers to focus on higher‑value problem‑solving. If you’re responsible for software quality, team efficiency, or large-scale code stewardship, this post reveals why agentic CI is quickly becoming the next strategic capability your organization can’t afford to ignore.

  • 🚢 Copilot metrics in GitHub Enterprise Cloud with data residency in public preview - Gain immediate clarity into how Copilot is being used across your enterprise—with granular metrics, dashboards, and APIs that unlock deeper insights for optimization, governance, and data‑driven decision‑making.

  • 🚢 Closing down notice of legacy Copilot metrics APIs - We are closing down three legacy Copilot metrics APIs as part of our transition to newer, more comprehensive usage metrics endpoints. Support for these APIs will be limited, and no new features will be developed.

  • 🚢 GitHub Changelog - Copilot - Skim through all of the recent Copilot changes.

CI/CD

Continuous Integration & Continuous Deployment with GitHub Actions. If you are involved in managing and authoring GitHub Actions workflows you'll want to dive into these updates to see how were are addressing enterprise needs in the areas of scalability, debugging, security and more.

  • 🚢 GitHub Actions: Smarter editing, clearer debugging, and a new case function - These updates give teams far greater clarity and control over complex workflow logic, making it faster and easier for enterprise engineering organizations to build, validate, and troubleshoot sophisticated automation at scale.

  • 🚢 GitHub Actions: Early February 2026 updates - This month, GitHub Actions introduces new capabilities, including custom runner autoscaling, expanded security controls for all users, and early access to new Windows and macOS runner images.

  • 🚢 arm64 standard runners are now available in private repositories - Teams can now accelerate native multi‑architecture CI workloads with new Linux and Windows arm64 standard runners in private repositories, delivering arm‑native performance without added infrastructure overhead.

  • 🚢 New fine-grained permission for artifact metadata is now generally available - GitHub is introducing a more precise artifact metadata permission that enterprise teams should adopt now to strengthen security controls and avoid upcoming workflow disruptions.

  • 🚢 Rate limiting for actions cache entries - Repositories pushing frequent cache uploads will want to understand this update because the new per‑repository upload rate limits can affect workflow reliability and may require action to avoid rejected cache entries.

  • 🚢 Docker and Docker Compose version upgrades on hosted runners - This update equips hosted runners with the newest Docker and Docker Compose capabilities while highlighting key workflow changes enterprises should evaluate to avoid disruptions.

  • 📢 Continuous AI in practice: What developers can automate today with agentic CI - Continuous AI introduces a new class of automation that tackles the judgment-heavy, context-dependent engineering work traditional CI can’t touch. It shows how lightweight, natural‑language‑driven agents can continuously handle tasks like documentation drift, dependency changes, performance regressions, and semantic bugs—freeing developers to focus on higher‑value problem‑solving. If you’re responsible for software quality, team efficiency, or large-scale code stewardship, this post reveals why agentic CI is quickly becoming the next strategic capability your organization can’t afford to ignore.

  • 🚢 GitHub Changelog - Actions - Skim through all of the recent security related changes. 

Security

Application security with GitHub, ensuring the code that lives in GitHub and the dependencies that go into the solutions you build are secure and do not contain any secrets.

Code Security

  • 📢 AI-supported vulnerability triage with the GitHub Security Lab Taskflow Agent - This post shows how AI‑powered taskflows can dramatically cut through the repetitive, error‑prone work of vulnerability triage—turning what used to take hours of manual auditing into a faster, more consistent, and more scalable workflow. For enterprise developers and engineering leaders, it demonstrates why adopting LLM‑driven automation isn’t just a productivity boost but a strategic advantage for reducing false positives, uncovering real vulnerabilities, and strengthening security posture at scale. If you’re building or overseeing modern software systems, you need to understand these patterns now because they point to the future of secure, AI‑assisted development.

  • 🚢 CodeQL 2.24.0 adds Swift 6.2 and .NET 10 support, and improves file handling for minified JavaScript - CodeQL 2.24.0 delivers important upgrades that expand coverage across Swift 6.2, .NET 10, modern JavaScript frameworks, and key security‑sensitive libraries—meaning your code scanning now detects more issues in the languages and stacks you rely on most. It also sharpens analysis accuracy across C#, Java, Rust, and C/C++, reducing false positives so teams can focus on real vulnerabilities rather than noise. 

Secret Protection

Supply Chain Security

Additional Security Updates

  • 📢 Community-powered security with AI: an open source framework for security research - This post introduces GitHub’s new open source Taskflow Agent framework—an AI‑powered, MCP‑driven system that lets teams encode and scale security expertise using natural language. For enterprises facing rising complexity and vulnerability volume, this framework shows how AI agents can operationalize security research, accelerate variant analysis, and make threat discovery repeatable across large codebases. If you care about strengthening your SDLC with transparent, customizable, and community-driven security automation, this is a roadmap you don’t want to miss.

  • 🚢 GitHub Changelog - Security - Skim through all of the recent security related changes. 

Engineering

An inside look at how we’re building the home for all developers. Resources based on our internal experiences. 

  • 📢 When protections outlive their purpose: A lesson on managing defense systems at scale - This post shows how even well‑intentioned, emergency‑driven protection mechanisms can silently become technical debt that impacts real customers—something every large-scale software platform is vulnerable to. It explains why continuous observability, lifecycle management, and regular reevaluation of safeguards are essential to prevent outdated defenses from degrading user experience. If you operate or influence complex systems, this is a sharp reminder that protecting your platform isn’t just about adding controls—it’s about knowing when to retire them.

Legend

That’s it for the February '26 edition of the MER. Follow GitHub Enterprise on LinkedIn to see when the next round of key updates become available. 

We want to hear from you! Did you find this curated list of updates from GitHub helpful? Do you have suggestions on how we can provide the information that is going to be the most useful and timely for your role? Provide your feedback in the GitHub Community: February ‘26 enterprise roundup.

Share