Anthropic · Coding agent (CLI, IDE extensions, desktop, web)

Claude Code

Anthropic's agentic coding tool that reads your codebase, edits files and runs commands from the terminal, IDE, desktop or browser.

AI Coding Assistants AI Agents & Frameworks

Provider: Anthropic
Latest confirmed model: Claude Opus 5.5 (Sep 22, 2026)
Pricing: Paid only
Free plan: No
API: Yes
Status: Generally available
Last verified: Oct 19, 2026

Visit official website Pricing page API docs

What is Claude Code?

Claude Code is an AI agent you give tasks in plain language; it plans the work, edits files across your project, runs commands and tests, and can create commits and pull requests. It runs in a terminal, in VS Code and JetBrains, in the Claude desktop app and on the web, and all surfaces share the same project settings. It is included with paid Claude plans or can be billed through an Anthropic API key.

Multi-step engineering tasks such as fixing bugs, writing tests or updating dependencies take many manual steps. Claude Code carries out those steps and reports back for review.

Why use Claude Code?

Practical benefits

Works across terminal, IDE, desktop, web and mobile
Project memory via CLAUDE.md, plus skills, hooks and MCP
Sub-agents and background agents for parallel work
Scheduled routines and CI integration
Agent SDK to build custom agents

Who it suits

Software engineers
Teams automating reviews and maintenance
Developers building custom agents

When another tool may fit better

Users on the free Claude plan
Non-technical users who want a visual app builder

Learning curve: Some experience helps
Skills needed: Comfort with a terminal or IDE and reviewing code changes.

Example workflows

Ask Claude Code to write tests for a module, run them and fix failures
Mention @Claude in Slack with a bug report and receive a pull request

What can it do?

Terminal CLINative installer for macOS, Linux, WSL and Windows; also Homebrew, WinGet and Linux package managers.
IDE extensionsVS Code (and Cursor) extension and a JetBrains plugin.
Desktop and web Depends on planCode tab in the Claude desktop app and claude.ai/code; desktop requires a paid subscription.
MCP, skills and hooksConnect external tools and package repeatable workflows.
Routines and scheduled tasksCloud routines and local scheduled tasks for recurring work.
GitHub Actions / GitLab CI and code reviewAutomate PR review and issue triage.
Agent SDKBuild custom agents using Claude Code's tools.

Good for

Bug fixing
Writing tests
Refactoring
Automated PR review
Dependency updates
Custom agent development

Where it runs

macOSWindowsLinux/WSLVS CodeJetBrains IDEsDesktop appWebiOS/Android (via Claude app)Slack

Pricing and plans

Paid only Claude Code is included in Claude Pro ($20/month, or $17/month billed annually), Max (from $100/month), Team and Enterprise plans. It can also be used pay-as-you-go with an Anthropic API key at standard token rates.

Plan	Price	Billing	Free trial / tier	Usage limits	Main inclusions	Model access	API access	Restrictions
Free (Claude)	Free	per month	Free tier	Everyday chat use	Claude chat	—	—	Does not include Claude Code
Pro	$20/mo $17/mo billed annually Annual billing is $200 up front	per month	—	At least 5x Free usage per 5-hour session	Claude Code included	—	Separate API billing	—
Max	$100/mo From $100/month; monthly billing only; 5x or 20x Pro usage tiers	per month	—	5x or 20x Pro usage per 5-hour session	Claude Code included	—	Separate API billing	Price of the 20x tier not captured
Team Standard seat	$25/mo $20/mo billed annually Per seat	per month	—	More than Pro	Claude Code included Team admin	—	—	—
Team Premium seat	$125/mo $100/mo billed annually Per seat	per month	—	5x Standard seat usage	Claude Code included	—	—	—
Enterprise	$20/yr $20 per seat per month billed annually, 20-seat minimum, plus usage billed at API rates	per year	—	Usage-based	Claude Code included	—	Usage at API rates	20-seat minimum

Prices in USD · US / global list price · Taxes may apply · Verified Oct 19, 2026

Worked example (usage-based API pricing): sending 1,000,000 input tokens and receiving 200,000 output tokens at the listed Claude Opus 5.5 rates ($4 and $20 per 1M) costs about $8 before tax. Real bills depend on the model, caching, batch discounts and any tier you qualify for.

Official pricing page (opens in a new tab)

API and developer access

Official API: Yes
Documentation: API docs (opens in a new tab)
Authentication: API key (ANTHROPIC_API_KEY) or Claude subscription login
Pricing model: per 1M tokens
Free API quota: None documented — the free plan above (if any) is for the app, not the API
Rate limits: Not documented
Context / input limits: Not documented
SDKs: Claude Agent SDK
Streaming: Not documented
Batch processing: Yes
Structured output: Not documented
Tool / function calling: Yes
Commercial use: See the provider terms
Data retention & privacy: See the provider privacy terms

Published API rates

Item	Price	Unit
Claude Opus 5.5 input	$4	per 1M input tokens
Claude Opus 5.5 output	$20	per 1M output tokens
Claude Sonnet 5.5 input	$2	per 1M input tokens
Claude Sonnet 5.5 output	$10	per 1M output tokens
Claude Haiku 5.5 input (<=100K tokens)	$0.10	per 1M input tokens
Claude Haiku 5.5 output (<=100K tokens)	$0.50	per 1M output tokens
Claude Fable 5.1 input	$10	per 1M input tokens
Claude Fable 5.1 output	$50	per 1M output tokens

A free Claude Code account is not the same thing as free API access. Unless a free API quota is listed above, API calls are billed separately.

Verified Oct 19, 2026

Advantages and disadvantages

Advantages

Many surfaces sharing one configuration
Strong extensibility (MCP, skills, hooks, SDK)
Choice of subscription or API billing

Disadvantages

Not included in the free Claude plan
Usage on subscriptions is limited per 5-hour session
Changes still need human review

Privacy: Terminal CLI, VS Code and JetBrains can also run through supported third-party cloud providers. Check Anthropic's data usage terms for your plan.

Real-world examples

A developer fixing a failing test suite

Install Claude Code and run claude in the project
Ask it to run the tests and fix failures
Review the diff and commit

needs paid plan

A CI pipeline reviewing every pull request

Add the Claude Code GitHub Action
Configure an Anthropic API key
Let it comment on new PRs

needs API key

A team lead scheduling a weekly dependency audit

Create a routine from the web or with /schedule
Describe the audit task
Review the report or PR it produces each week

needs paid plan

Claude Code alternatives

OpenAI Codex

OpenAI's coding agent, included with ChatGPT plans and usable with an API key.

has a free plan · offers an API

Claude Code vs OpenAI Codex

GitHub Copilot

An AI pair programmer and coding agent that works in your editor, terminal and on GitHub itself.

has a free plan · no public API

Claude Code vs GitHub Copilot

Cursor

An AI-first code editor with local and cloud agents, now sold on a shared Cursor and Grok subscription.

has a free plan · offers an API

Claude Code vs Cursor

Devin Desktop (formerly Windsurf)

The renamed Windsurf editor, rebuilt around managing local and cloud coding agents.

has a free plan · offers an API

Claude Code vs Devin Desktop (formerly Windsurf)

Amazon Q Developer

AWS's AI coding assistant, whose IDE plugins reach end of support on April 30, 2027 with Kiro as the recommended replacement.

has a free plan · no public API

Claude Code vs Amazon Q Developer

Compare Claude Code with these side by side

Frequently asked questions

Is Claude Code free?

No. The free Claude plan does not include it. It comes with Pro, Max, Team and Enterprise plans, or you can pay per token with an API key.

How much does Claude Code cost?

The cheapest option is Claude Pro at $20 per month ($17 per month billed annually). Max starts at $100 per month for 5x or 20x Pro usage.

Where can I use Claude Code?

In a terminal, VS Code, JetBrains IDEs, the Claude desktop app, the web at claude.ai/code, and via Slack and CI pipelines.

Can I use it with an API key instead of a subscription?

Yes. With an Anthropic Console account, usage is billed at API token rates, for example $4 input and $20 output per million tokens for Opus 5.5.

Can I build my own agents with it?

Yes. The Agent SDK lets you build custom agents using Claude Code's tools with control over orchestration and permissions.

Does it work on Windows?

Yes. There are native installers for PowerShell and CMD, plus WinGet; Git for Windows is recommended so it can use Bash.