Claude Code

What it does

Claude Code works directly inside your project: it reads the codebase, edits several files at once, runs tests and shell commands, and creates commits and pull requests. You can use it in the terminal, as an extension for VS Code and JetBrains, in the desktop app or in the browser, where tasks keep running in a cloud environment even when your computer is off. MCP lets you connect further tools such as ticket systems, and a CLAUDE.md file holds your project rules. Several agents can work on subtasks in parallel, and recurring tasks can be scheduled.

Who it suits

Developers and teams who are comfortable with Git and the command line and want to hand off routine work such as tests, refactoring, dependency updates or code reviews.

Data protection for businesses

In Pro and Max you decide in the claude.ai privacy settings whether your data is used for training; if enabled, Anthropic keeps it for up to five years, otherwise for 30 days. In Team, Enterprise and via the API, Anthropic does not train on your code. The data processing agreement with standard contractual clauses is part of the commercial terms and does not cover Pro and Max. Anthropic stores data in the US; to have it processed in the EU, run Claude Code through Google Cloud with an EU region, for example.

How we use it

Claude Code is part of our everyday development, for example for refactoring, tests and documentation. We review what Claude Code proposes before it goes into a project.

Limits

Not included in the free Claude plan. Claude Code shares its usage allowance with the chat. The agent runs real commands on your system, so review changes before accepting them.