Claude Code is Anthropic's agentic coding tool — a command-line AI assistant that reads your codebase, edits files, runs commands, and works directly inside your existing development workflow. Rather than a simple autocomplete or chat-based helper, it acts as an agent: given a task in plain language, it plans an approach, searches and edits across multiple files, runs tests, and can open pull requests on its own, asking permission before it changes files or executes commands.
Built by Anthropic
Claude Code is a product of Anthropic, the company behind the Claude family of AI models. Anthropic describes itself as an AI safety and research company built as a Public Benefit Corporation, with a stated purpose of "the responsible development and maintenance of advanced AI for the long-term benefit of humanity." Claude Code is the company's dedicated tool for software engineering work, sitting alongside Anthropic's broader Claude product line.
Core features
Claude Code can map and explain an entire codebase in seconds using agentic search, so developers don't need to manually select files for context. It handles both routine tasks — bug fixes, writing tests, resolving merge conflicts, updating dependencies — and larger jobs like refactors and new feature implementation. It integrates with git, GitHub, and GitLab to read issues, write code, run tests, and submit pull requests, and it connects to external tools and data sources through the open Model Context Protocol (MCP), letting it pull from things like Google Drive, Jira, or Slack.
Teams can customize its behavior with a CLAUDE.md file that stores project-specific instructions, plus reusable "skills" and "hooks" that run shell commands around its actions. Multiple Claude Code agents can run in parallel on different parts of a task, and Anthropic's Agent SDK lets developers build fully custom agents on top of the same underlying engine. For automation, the CLI supports a headless mode (claude -p) that pipes into scripts, CI pipelines, GitHub Actions, or GitLab CI/CD, and scheduled "routines" can run on a timer or in response to events. Claude Code is available across the terminal (macOS, Linux, Windows, WSL), a VS Code extension, JetBrains plugins, a standalone desktop app, a web version, and mobile access via remote control — all sharing the same settings, memory, and MCP connections.
Pricing
Claude Code is not available on Anthropic's free plan. It's included with a paid Claude subscription: the Pro plan starts at $17/month billed annually (or $20/month billed monthly), and Max plans start at $100/month. Team plans include Claude Code with a standard seat priced from $20/month (annual) or $25/month (monthly), and a premium seat from $100/month (annual) or $125/month (monthly); Enterprise pricing is available on request. Developers and organizations can also access Claude Code through an Anthropic Console account, billed at standard API usage rates rather than a flat subscription.
Who it's for
Claude Code is built for developers and engineering teams who want an AI collaborator that works directly in their existing tools — terminal, IDE, git, and CI — rather than a separate chat window. It suits individual developers automating tedious maintenance work, teams wanting automated PR review and issue triage, and organizations building custom AI-driven engineering workflows on top of the Agent SDK.