Claude Code

Claude Code is Anthropic’s agentic coding tool. It reads your codebase, edits files, runs commands and tests, and works through multi-step tasks while you review the changes. You can use it in the terminal, in supported code editors, on the web and in desktop apps. Unlike most tools in this directory, Claude Code has no separate price: it comes with a paid Claude plan.

Quick facts

• Made by: Anthropic, the company behind Claude
• Free plan: No. The free Claude plan does not include Claude Code.
• How you get it: Included with Claude Pro, Max, Team and Enterprise plans, or by paying per token through the Anthropic API
• Works in: The terminal, supported IDEs such as VS Code and JetBrains, the web and desktop
• Pricing last checked: October 2026, using several recent published sources. Usage limits and plan names change often, so confirm on Anthropic’s official pricing page.

What Claude Code is best at

• Working across a whole project: understanding how files fit together, rather than only completing the line you are typing.
• Multi-step tasks: planning a change, editing several files, running tests and fixing failures.
• Explaining unfamiliar code: asking questions about a codebase in plain language.
• Automation: scripting and headless use for tasks such as code review or migrations in a pipeline.
• Using your existing tools: it runs in your terminal alongside git, test runners and build tools.

Plans and pricing

On Pro and Max, Claude Code and the Claude apps draw from the same quota, made up of a rolling five-hour session limit plus a weekly limit. Subscribers can enable usage credits to keep working past the included limit. Running Claude through the Agent SDK or headless pipelines was reported to draw from a separate monthly credit instead of your interactive limits. Which models you can use differs by plan, and the exact limits change, so confirm on Anthropic’s official page.

Best for

• Developers who want an agent that works in their terminal and across a codebase.
• Teams that already use Claude and want coding included in the same subscription.
• People who want to script or automate coding tasks.

Less suited to

• Anyone who wants a free coding tool, since there is no free Claude Code plan.
• Developers who prefer a full visual editor built around AI, such as Cursor or Windsurf.
• Users who need fixed, unlimited usage, because limits apply on every subscription.

How to get started

1. Subscribe to a paid Claude plan, or set up an API key if you prefer pay-per-token billing.
2. Install Claude Code in your terminal or editor and sign in with the same account.
3. Start in a small project with a clear task, and read every change before you accept it.
4. Use the built-in usage commands to see where you are against your limits.
5. Move up a plan only if you regularly hit your limits.

Prompt ideas to copy

Explain how authentication works in this project. List the files involved and point out anything that looks risky.

Add input validation to the signup endpoint. Follow the existing patterns, write tests for the new cases and run them. Summarise what you changed.

Find and fix the cause of this failing test: [paste error]. Make the smallest change that works and explain why it failed.

Refactor this module to remove duplication without changing behaviour. Run the test suite after each step and stop if anything breaks.

Limitations and things to watch

• Agent changes can be wrong or insecure, so review diffs and run your own tests.
• Usage limits are shared with the chat apps, and heavy sessions can use them quickly.
• Plan names, limits and which models are included change often.
• Check your organisation’s rules before pointing any agent at private or licensed code.

Alternatives

Compare GitHub Copilot for AI inside your current editor, Cursor and Windsurf for AI-first editors, and Replit for building and hosting apps in the browser.