Introducing Claude Code
Published by AnthropicWatch on YouTube
Updated Oct 4, 2026
AI Coding Tools
Claude Code
Anthropic's coding agent for delegating whole code changes
From $17/mo
Claude Code at a glance
from our research
Anthropic's coding agent, separate from the Claude chat app, for developers in a terminal or IDE
Plans a change, edits files across the codebase, runs the test suite and repairs failures
Shares Claude's five-hour and weekly allowance, and Opus draws it down faster than Sonnet
Pro costs $20 a month, or $17 a month billed yearly, and covers both Claude Code and chat
Overview
What is Claude Code?
Claude Code is a coding agent from Anthropic, the American company founded in 2021 that also makes the Claude chat app; the two are separate products sold through the same subscriptions. You start Claude Code inside a project and give it a task in plain language, such as fixing the failing tests or adding a feature. It then works in a loop: it reads the relevant files, edits code across as many files as needed, runs commands such as the test suite, checks the results and repeats until the task is done. You can interrupt at any point, and it can also commit changes and open pull requests.
Before it edits a file, Claude Code saves a checkpoint you can rewind to. A permission mode sets how much it does without asking: Manual asks before edits and shell commands, Plan only proposes an approach, and Auto lets a classifier block risky actions instead of prompting you. A CLAUDE.md file holds project conventions it reads every session, and MCP servers connect it to tools such as Jira, Slack and Google Drive.
It runs in the terminal on macOS, Linux and Windows, in VS Code (and forks such as Cursor) and JetBrains IDEs, in the Claude desktop app, and in the browser at claude.ai/code, where tasks run on Anthropic's cloud machines. It is included with the Pro, Max, Team and Enterprise plans but not Free, and shares one usage allowance with the chat app; developers can also pay per token with an API key.
Our verdict on Claude Code
Updated Oct 4, 2026Pricing checked Oct 3, 2026
Claude Code suits developers who would rather hand over a whole task than accept suggestions line by line. Anthropic contrasts it with inline code assistants that only see the current file: it reads across the repository, makes coordinated edits, runs the tests and fixes what fails, which fits bug fixes, test coverage, refactors and dependency upgrades. The same engine runs on every surface, so a team's CLAUDE.md, settings and MCP servers carry across them, and the command-line version can be scripted or run in CI.
The trade-off is review work and cost control. An agent that edits many files produces large changes to read, and checkpoints only cover local file edits; anything it does to a database, an API or a deployment cannot be rewound, so the permission settings matter. On subscriptions, coding and chat draw on the same five-hour and weekly limits, Opus uses noticeably more of them than Sonnet, and long sessions grow costly because the whole conversation is resent on every turn. Some features also depend on the plan: automatic Code Review on pull requests is for Team and Enterprise only.
Most individual developers will start on Pro at $20 a month, or $17 a month billed yearly ($200 up front), which includes Claude Code alongside the chat app. Heavy users can move to Max 5x at $100 a month or Max 20x at $200 a month, billed monthly, for five or twenty times Pro's per-session usage. Teams pay $25 per Standard seat a month, or $20 billed yearly, for 1.25 times Pro's usage; Premium seats at $125, or $100 billed yearly, carry five times a Standard seat's usage. Enterprise charges $20 per seat a month, billed annually, plus all usage at API rates, so its bill grows with how much each developer runs the agent.
Choose Claude Code if you work in a terminal or IDE, can describe changes clearly and will review the diffs. If you want suggestions as you type, GitHub Copilot Pro at $10 a month includes unlimited inline suggestions. If your team already pays for ChatGPT, Plus at $20 a month includes OpenAI's Codex agent in the terminal, the IDE and the browser.
Written by Lokesh Kapoor
Pros & cons
Claude Code pros and cons
What it does well
- Runs in the terminal, VS Code, Cursor, JetBrains IDEs, the desktop app and the browser
- Can be scripted and run in CI through GitHub Actions or GitLab CI/CD
- Checkpoints rewind file edits, and permission modes control what it does unasked
- Team, Enterprise and API code isn't used for training unless the customer opts in
Where it falls short
- Not part of Claude's Free plan; access starts with Pro or a pay-per-token API key
- Checkpoints can't undo changes it makes to databases, APIs or deployments
- Pro and Max plans don't get automatic Code Review on pull requests
Editor's assessment from vendor documentation and published pricing
Pricing
Claude Code pricing
Claude Code's paid plans start at $17/month, below the $19 median across the 14 tools in our AI Coding Tools ranking.
Pro
$17/mo
per month
or $20/mo billed monthly
For one person
- Claude Code
- More usage than Free
- Projects
- Claude in Chrome and Microsoft 365
Usage: At least 5x Free's usage per five-hour session, plus a weekly limit. The yearly plan is $200 billed up front. Extra usage at API rates if you turn on usage credits.
Max 5x
$100/mo
per month
For one person
- 5x Pro's usage per five-hour session
- Claude Code
- Higher output limits
- Priority access at busy times
Usage: Weekly limit across all models. Monthly billing only.
Max 20x
$200/mo
per month
For one person
- 20x Pro's usage per five-hour session
- Claude Code
- Higher output limits
- Priority access at busy times
Usage: Weekly limit across all models. Monthly billing only.
Team
$20/mo
per seat, per month (Standard seat)
or $25/mo billed monthly
Minimum 2 seats
- Claude Code with every seat
- More usage than Pro
- Central billing and admin
- SSO
- Enterprise search
Usage: Each Standard seat gets 1.25x Pro's per-session usage plus a weekly limit. 2 to 150 seats; Standard and Premium seats can be mixed.
Team Premium
$100/mo
per seat, per month (Premium seat)
or $125/mo billed monthly
Minimum 2 seats
- 5x a Standard seat's usage
- Claude Code
- Mix with Standard seats
Usage: 6.25x Pro's per-session usage plus a weekly limit.
Enterprise
$20/mo
per seat, per month, billed annually, plus usage at API rates
Minimum 20 seats
- All Team features
- Spend limits per user and organisation
- SCIM, audit logs and Compliance API
- Custom data retention
Usage: The seat fee covers access only: all usage is billed at standard API rates, with no included allowance. Self-serve needs 20+ seats; sales-assisted 50+. Yearly billing only.
Checked against Anthropic's pricing page on October 3, 2026 · see the full breakdown
Capabilities
Claude Code features
Checked against the same feature list as every other tool in its category.
Platform
- Not included:Built-in hosting & deployRuns and deploys the application without configuring separate infrastructure.
Team & Governance
- Included:No training on your codeBusiness tiers exclude your source from model training by default.
- Included:Self-hosted or VPC optionRuns inside your own infrastructure for regulated or air-gapped environments.Via your AWS, Google Cloud or Azure account
- Included:Team admin & SSOCentral seat management, policy controls and single sign-on.
Code Generation
- Included:In-editor chatAsk questions about the code you are looking at without leaving the editor.
- Not included:Inline autocompleteMulti-line suggestions as you type, accepted with a keystroke.
- Included:Test generationWrites unit and integration tests against existing code.
Codebase Awareness
- Included:Custom project rulesPersistent instructions describing your conventions, applied to every request.
- Included:Tool & MCP integrationConnects to external tools and data sources so the assistant can read docs, tickets or a database.
- Included:Whole-repository indexingIndexes the entire codebase so suggestions follow your existing patterns and reuse your helpers.Through on-demand search, not a stored index
Agent Autonomy
- Included:Automated PR reviewReviews pull requests and comments on problems before a human does.
- Included:Background & async agentsWorks on a task on its own while you do something else, then reports back.
- Included:Multi-file editsApplies a single instruction coherently across several files at once.
- Included:Runs commands & testsExecutes builds and test suites, reads the output and fixes what it broke.
Ecosystem
Claude Code integrations
Videos
Watch Claude Code in action
Product walkthroughs from Claude Code's own YouTube channel.
Mastering Claude Code in 30 minutes
Published by AnthropicWatch on YouTube
Alternatives
Top Claude Code alternatives
Tools that do the same job, from the same category.
Fit check
Will Claude Code fit your team?
Support
- Email support
- Knowledge base
Training
- Documentation
- Video tutorials
Deployment
- macOS
- Windows
- Linux
- Web
Typical users
- Freelancers
- Small businesses
- Mid-size companies
- Enterprises
FAQs
Claude Code FAQs
Written by our editorial team.
Is Claude Code free?
No. Claude Code is not part of Claude's Free plan. The cheapest way in is Pro at $20 a month, or $17 a month billed yearly, which covers both Claude Code and the Claude chat app. Max, Team and Enterprise plans also include it. Developers can instead sign in with an Anthropic Console API key and pay per token, with no subscription. See Claude Code pricing for every plan.
How do Claude Code's usage limits work?
On Pro and Max, Claude Code shares one allowance with the Claude chat app: a per-session limit that resets every five hours, plus a weekly limit. Max 5x and Max 20x give five and twenty times Pro's per-session usage. Opus uses much more of the allowance than Sonnet, and every turn resends the whole conversation, so clearing between tasks stretches it. When you run out, you can wait for the reset, upgrade, or turn on usage credits billed at API rates.
Does Anthropic train its models on your code?
Not on Team, Enterprise or API accounts: under commercial terms Anthropic does not train on code or prompts sent to Claude Code unless the customer opts in. On Pro and Max, a privacy setting decides it; with model improvement on, sessions can be used for training and are kept for five years, and with it off they are kept for 30 days. Claude Code also stores session transcripts on your machine for 30 days by default.
Which editors and platforms does Claude Code support?
Claude Code runs in the terminal on macOS, Linux and Windows, in VS Code and VS Code forks such as Cursor, and in JetBrains IDEs including IntelliJ, PyCharm and WebStorm, where the plugin needs the command-line tool installed. It also runs in the Claude desktop app, in the browser at claude.ai/code, and from the Claude iOS and Android apps for cloud tasks, plus GitHub Actions and GitLab CI/CD. For an editor-first alternative, see Cursor vs Claude Code.
Can Claude Code change files without asking?
Only if you let it. In Manual mode it asks before every file edit and shell command, Accept edits lets it change files but still asks about other commands, Plan mode only proposes changes, and Auto mode has a classifier block risky actions instead of prompting you. You can pre-approve trusted commands such as running tests in a settings file. File edits can be rewound with checkpoints, but changes to databases, APIs or deployments cannot.
Sources & methodology
Product details come from Anthropic's public website and documentation (United States). Every plan and price was checked against the vendor's pricing page on October 3, 2026. We haven't tested this product hands-on; the verdict is a desk assessment. How we research and rank tools
- Founded 2021
- 1 language supported
- anthropic.com
Keep researching