Claude Code

Claude Code is Anthropic’s agentic coding tool that runs directly in your terminal, designed for developers who want AI assistance embedded in their existing command-line workflow without switching to a browser or IDE plugin. It’s Claude with full visibility into your project: it can read files, write code, run commands, search the codebase, and execute multi-step engineering tasks autonomously.

The core workflow is open a terminal in your project directory, run claude, and describe what you need. Claude Code reads relevant files to understand context, executes the task, runs tests or linters to verify correctness, and reports back. For large tasks it breaks them into steps, showing its work at each stage.

Claude Code has first-class MCP support. Add MCP servers to its configuration and Claude Code gains access to those tools during coding sessions — query a database to understand your data model while writing a migration, call the GitHub API to check open issues while fixing bugs, or browse documentation while implementing a new library integration.

The --print flag enables non-interactive use: pipe Claude Code into scripts, CI pipelines, or other automation tools. This makes it possible to build workflows like automatically generating tests for new files on commit, or having an AI code review stage that runs in your CI system.

Claude Code is available on claude.ai/download and requires a Claude.ai Pro or Team subscription. Usage within Claude Code counts against your plan’s conversation limits. Enterprise customers can access Claude Code with custom rate limits and data privacy controls.