Anthropic is preparing a web version of Claude Code, expanding beyond the current workflow tied to desktop or API use. The feature is designed to resemble OpenAI’s Codex on the web, giving developers direct access to a coding agent through a browser interface. The new section will allow GitHub integration, letting users connect repositories, generate tasks, and run them inside isolated environments before finalizing the results. This setup points toward managed sandboxes where Claude Code can execute and test code safely without requiring local setup.
Research preview of Claude Code on Claude[.]ai
— Tibor Blaho (@btibor91) August 26, 2025
- updated layout with the prompt composer and sessions moved to the sidebar
- repository setup requires installing the GitHub Claude app on your repository and committing the "Claude Dispatch" GitHub workflow file
- you can enable… https://t.co/GdWctXa2oK pic.twitter.com/8WOx95mMrL
The beneficiaries will be developers and teams who prefer browser-based tools and want faster onboarding without complex configuration. It also positions Claude Code as a competitor to existing cloud IDEs that are increasingly embedding AI copilots. While no release date is confirmed, the pace of development suggests a possible rollout in the coming months.
The addition fits Anthropic’s broader strategy of making Claude not just a conversational model but also a workspace with specialized agents. With GitHub connectivity and automated environments, this version could lower friction for coding projects and align Claude Code with a growing trend of AI-driven, cloud-hosted engineering assistants. If deployed as expected, it would strengthen Anthropic’s offering for both individual developers and enterprise teams looking for integrated AI coding workflows.
