Anthropic’s Claude Code is now available on the web for its Pro and Max users

Claude Code is Anthropic’s AI coding assistant designed to help developers fix bugs, refactor code, or even write complete programs from scratch. Until now, it’s been something you’d typically access through tools like Visual Studio.

That’s changing, though. Anthropic has now made Claude Code available on the web, letting users run coding tasks directly from their browser. It’s a big step toward accessibility, allowing you to use Claude Code anywhere without needing a local setup.

Right now, the feature is in research preview and limited to Pro and Max subscribers, with support for Team and Enterprise plans expected later on.

How Claude Code on the web works

Developers can connect Claude Code to their GitHub account, allowing it to securely access their repositories. Once linked, Claude will clone the repository to an Anthropic-managed virtual machine, set up a coding environment, and begin running tasks automatically.

Right now, Claude Code on the web can already analyze, modify, and push code, as well as create a pull request when it’s finished. The virtual machine comes with several programming languages pre-installed, including Python, Node.js, Java, Rust, and C++, making it a flexible option for most workflows.

Anthropic also notes that network access is tightly controlled for security reasons, with only trusted domains allowed by default. Users can switch seamlessly between Claude Code’s web environment and their local setup, creating a versatile workflow for developers on the go.

Anthropic’s expanding ecosystem and Microsoft partnership

Anthropic appears to be in a strong position, and this move makes Claude even more accessible to users. As one of OpenAI’s main competitors, Anthropic has also partnered with Microsoft to bring Claude to Azure as part of its growing AI ecosystem. Claude is now integrated across Outlook, Teams, and OneDrive, extending its reach into everyday productivity tools.

This steady expansion shows Anthropic’s increasing focus on enterprise and developer solutions. With Claude Code now available on the web, it’s another step toward making Claude a full development assistant — something you can use from your phone, browser, or preferred coding environment.

A major step toward accessible AI development

Whether you’re big on AI or not, or whether you think the AI bubble will eventually burst, for now, AI is here to stay—and Anthropic’s latest step makes Claude Code far more accessible to those who use it.

This web rollout bridges the gap between local coding tools and cloud-based AI agents, which is ideal for anyone working remotely or without a high-powered machine. Alongside accessibility, Anthropic also places heavy emphasis on security. Each session runs in an isolated virtual machine with tightly controlled network access, and sensitive credentials never leave the secure proxy used for authentication.

When you start a task on Claude Code on the web, the process is fairly simple. Claude clones your GitHub repository to a secure virtual machine, sets up the coding environment, and configures internet access based on your preferences. From there, it analyzes your code, makes the necessary changes, runs tests, and checks its own work. Once it’s done, you’re notified and can review everything before creating a pull request.

There are still a few limitations to note. Claude Code on the web currently only works with GitHub repositories, and moving sessions between web and local setups requires authentication to the same account. It also shares rate limits with your overall Claude usage, so running several tasks in parallel will use them up faster.

Even with these restrictions, Anthropic’s cautious, security-first approach shows it’s serious about making AI coding both powerful and responsible—a strong foundation for whatever comes next, whether you buy into all this stuff or not.

FAQ

Follow Windows Central on Google News to keep our latest news, insights, and features at the top of your feeds!