Last updated: October 29, 2026
Claude Code is Anthropic’s agentic coding tool for the terminal, the Claude desktop app and VS Code. In October 2026 it added mods, code plugins that can change its interface, and Code Review became generally available. It defaults to Opus 5.5 and is included in every paid Claude plan, starting with Pro at $20 a month.
Claude Code reads your codebase, edits files, runs commands and works through multi-step tasks. You can sign in with a Claude Pro, Max, Team or Enterprise account, a Claude Console account, or a cloud provider such as Amazon Bedrock or Microsoft Foundry (quickstart). The latest version on GitHub releases is v2.1.312, published October 27, 2026. It runs on Anthropic’s Claude models, such as Claude Haiku 5.5.
Contents
- What’s new in Claude Code (October 2026)
- How to install Claude Code
- How to use Claude Code mods
- How to set up Claude Code Review
- Opus 5.5 vs Sonnet 5.5 in Claude Code
- Auto mode and /goal
- Claude Code pricing
- Limitations to know
- Frequently asked questions
What’s new in Claude Code (October 2026)
| Feature | Date or version | What it does |
|---|---|---|
| Mods | October 1, 2026 | Code plugins that rewrite events and draw interface elements |
| Code Review GA | October 15, 2026 | Automated pull request review for Team and Enterprise, about 3x faster |
| Opus 5.5 default | v2.1.280+ | The default setting resolves to Opus 5.5 on Pro, Max, Team, Enterprise and the API |
| Sonnet 5.5 | v2.1.284+ | Selected by the sonnet alias on the Anthropic API |
| Auto mode default | v2.1.283+ | Auto mode is the starting permission mode in the terminal and VS Code |
| ultrathink to xhigh | v2.1.311+ | Typing ultrathink raises that turn to xhigh effort |
How to install Claude Code
The quickstart gives these steps:
- Run the native installer. On macOS, Linux or WSL, run
curl -fsSL https://claude.ai/install.sh | bash. On Windows PowerShell, runirm https://claude.ai/install.ps1 | iex, or usewinget install Anthropic.ClaudeCode. - Check the install with
claude --version. - Start a session in your project with
cd your-project && claude, then sign in when prompted. - Learn the core commands:
claude -p "query"for a one-off answer,claude -cto continue the last conversation andclaude -rto resume an earlier one. Press Shift+Tab to cycle permission modes. - Keep it current with
claude update, since the newest models and features need recent versions.
How to use Claude Code mods
Anthropic introduced mods on October 1, 2026. A mod is a set of TypeScript or JavaScript functions inside a plugin that Claude Code calls in its own process. A mod can rewrite a prompt, hold or change a tool call, add a /command that runs your function without a Claude turn, and draw panes, buttons and bands above the prompt. See the mods overview for details.
- Update. Mods need Claude Code CLI v2.1.287 or later, or Desktop v2.1.286 or later.
- Install a mod from a marketplace you trust with
/plugin install token-chart@your-orgin a session. If a session is already open, run/reload-plugins. - Or ask Claude to write one, for example “make a mod that shows the current git branch above the prompt.” Claude uses its built-in
plugin-authoringskill. - Confirm it loaded with
/pluginand the Installed tab.
How to set up Claude Code Review
Code Review became generally available on October 15, 2026 for Team and Enterprise plans. Anthropic says it is about 3x faster than before, with reviews that took about 20 minutes now taking 6 to 10. It leaves inline comments and a “Claude Code Review” check run, and it never approves or blocks a pull request. The setup guide gives these steps:
- Turn it on. An Owner goes to Admin settings > Claude Code > Code review and clicks Allow code review.
- Connect GitHub and install the Claude GitHub App on the organization, choosing which repositories it can access.
- Pick a level and trigger. Choose Quick, Standard (the default) or Deep. Reviews can run once after a pull request is created, after every push or only when requested.
- Test it by opening a pull request that is not a draft. A check run should appear within a few minutes.
On a pull request, comment @claude review for one review, @claude review always to also review future pushes, or @claude review stop to stop. Put review-only rules, such as “report at most five nits”, in a REVIEW.md file at the repository root, and keep project-wide context in CLAUDE.md.
Opus 5.5 vs Sonnet 5.5 in Claude Code
According to the model configuration docs, the default setting resolves to Opus 5.5 on Pro, Max, Team, Enterprise and Anthropic API accounts.
| Alias | Model on the Anthropic API | Minimum version |
|---|---|---|
opus |
Opus 5.5 | v2.1.280 |
sonnet |
Sonnet 5.5 | v2.1.284 |
haiku |
Haiku 5.5 | v2.1.293 |
opusplan |
Opus for planning, Sonnet for execution | — |
Switch models with /model, claude --model sonnet or the ANTHROPIC_MODEL environment variable. Set effort with /effort; the levels are low, medium, high, xhigh and max, and the 5.5 models default to medium. Type ultrathink to raise a single hard turn to xhigh.
Auto mode and /goal
Auto mode. Since v2.1.283, auto mode is the default starting mode. A separate classifier reviews each action and blocks anything that goes beyond your request or appears to be driven by hostile content (permission modes). To start in Manual mode instead, run claude --permission-mode default.
/goal. Use /goal to set a completion condition, such as /goal all tests in test/auth pass and the lint step is clean. Claude keeps taking turns until a separate model confirms the condition is met. Write the condition so Claude’s own output can prove it, because the evaluator does not run commands.
For a competing release from Google, see our Gemini 4 Argon guide.
Claude Code pricing
Anthropic’s pricing page says Claude Code is included in all paid plans and shares their usage limits.
| Plan | Price | Claude Code |
|---|---|---|
| Free | $0 | Not included |
| Pro | $20/month, or $17/month billed annually | Included |
| Max | From $100/month | Included |
| Team, standard seat | $25/month, or $20/month billed annually | Included |
| Team, premium seat | $125/month, or $100/month billed annually | Included |
| Enterprise | $20/seat/month billed annually, plus usage at API rates | Included |
| Code Review | Token usage billed through usage credits | Team and Enterprise |
Eligible Enterprise organizations can claim at least $10,000 in Code Review credits by January 15, 2027 (Anthropic). For Anthropic as a company, see our article on Anthropic’s IPO ticker.
Limitations to know
- Mods are not sandboxed and run with your permissions, so install them only from authors you trust. Use
--safe-modeto turn them off for one session. - Code Review is not available to organizations with Zero Data Retention, HIPAA or CMEK, and it works with GitHub.
- Auto mode reduces prompts but, as Anthropic warns, does not guarantee safety.
- If Claude Code seems unresponsive, check our AI outage tracker before you troubleshoot locally.
Frequently asked questions
Is Claude Code free?
No. Claude Code is not included in the Free plan. It is included in all paid plans, starting with Pro at $20 a month, or $17 a month billed annually. You can also use it with a Claude Console account that uses pre-paid API credits.
What are Claude Code mods?
Mods are TypeScript or JavaScript functions shipped inside a plugin that Claude Code runs in its own process. They can rewrite prompts and tool calls, add slash commands and draw interface elements such as panes. Mods require CLI v2.1.287 or later and are not sandboxed.
Which model does Claude Code use by default?
On Pro, Max, Team, Enterprise and Anthropic API accounts, the default setting resolves to Opus 5.5 (since v2.1.280). You can switch with /model or claude –model sonnet, which on the Anthropic API selects Sonnet 5.5.
What is Claude Code auto mode?
Auto mode lets Claude work without routine permission prompts while a separate classifier model reviews each action and blocks anything that goes beyond your request. Since v2.1.283 it is the default starting mode in the terminal and VS Code.
How much does Claude Code Review cost?
Code Review has no flat price. It is billed on token usage through usage credits, and the cost depends on pull request size and the review level you choose (Quick, Standard or Deep). It is available on Team and Enterprise plans.
Sources
- Claude Code docs: Quickstart
- Claude Code docs: Mods overview
- Anthropic: Code Review is now generally available
- Claude Code docs: Model configuration
- Claude pricing
Image: UK Secretary of State Peter Kyle (right) visits Anthropic in San Francisco, October 25, 2024. Picture by Alecsandra Dragoi / Department for Science, Innovation and Technology, CC BY 2.0, via Wikimedia Commons. Resized.
.jpg){kind=link}
Related guide: GitHub Copilot Fusion power-user guide: effort levels, billing and setup.
3 thoughts on “Claude Code: Power-User Guide to Mods, Code Review, Models and Pricing”