What Is Claude Code? Installation (Windows, Mac), Commands, Is It Free, Pricing, Limits, Permissions and Security
You type `claude` in a terminal inside your project folder and give Claude Code tasks in plain language; it reads and changes files and runs commands and tests. It's Anthropic's coding tool; it also works in VS Code, JetBrains, the desktop app and the browser, and installation is a single command. It isn't on the Free Claude plan: you need Pro, Max, Team, Enterprise or a pay-as-you-go Console (API) account. By default it carries out file edits and commands on its own (auto mode); if you want it to ask for approval, switch to Manual mode with Shift+Tab.
- What it does
- Reads and changes code, runs commands and tests in your project folder
- Turkish
- Replies: yes; interface language unclear
- Where
- Terminal, VS Code, JetBrains, desktop, web
- Current version
- 2.1.300 (October 14, 2026)
- Default permissions
- Auto mode
- Free?
- No; paid plan or API
- Price
- Pro $20
What Is Claude Code For? Can I Use It Without Coding?
Claude Code isn't a chat window that writes code; it's a tool that works in the project folder on your computer: it reads and changes files, runs commands and tests, and handles Git operations. You describe what you want in a plain sentence ("block empty submissions on the login form"), and it finds the relevant files and shows you the change.
Claude Code is a coding tool; if you don't have a code project, it won't help you. For tasks like summarizing documents or preparing Excel files or presentations, Claude's chat app is a better fit.
Claude guide: chat, file preparation and Cowork
If you have a code project but the terminal intimidates you, there are three ways in: the Code tab in the desktop app (no terminal needed, requires a paid subscription); claude.ai/code in the browser (no installation, but a GitHub repository is required); or the official terminal guide (it shows how to open a terminal and paste a command). If a developer or agency writes your site's code, they use Claude Code and you see the result.
Who it's for: anyone who works with code: a business building its own site or internal tool, an agency with developers, a software engineer. For someone who doesn't write code, the chat app is enough to start with.
How Do You Install Claude Code? Windows, Mac and Linux Commands
Installation is a single command: open a terminal and paste the line for your operating system. The command shows no progress while downloading; when it finishes, open a new terminal window. You need a Claude subscription (Pro, Max, Team, Enterprise) or a Console account first; the Free Claude plan doesn't work.
› curl -fsSL https://claude.ai/install.sh | bashThis method updates itself in the background. WSL is the Windows feature for running Linux inside Windows; you install Claude Code inside the WSL window, not from PowerShell.
› irm https://claude.ai/install.ps1 | iex› curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd- You don't need to run it as administrator.
- Git for Windows is optional: if it's installed, Claude Code uses its Bash tool (Git Bash); if not, it runs commands through PowerShell.
- If you get the error
The token '&&' is not a valid statement separator, you're running the CMD command in PowerShell; if you get'irm' is not recognized, you're running the PowerShell command in CMD.
!The sandbox isn't supported on native Windows; it is supported inside WSL 2. If you want to limit how much of the file system commands can reach on their own, WSL 2 is the better choice.
› brew install --cask claude-codeThere are two release channels: claude-code is the stable channel (usually about a week behind, skipping releases with major bugs), and claude-code@latest is the newest channel. A Homebrew installation doesn't update itself: run brew upgrade claude-code (or brew upgrade claude-code@latest).
› winget install Anthropic.ClaudeCodeA WinGet installation doesn't update itself either: run winget upgrade Anthropic.ClaudeCode from time to time.
› npm install -g @anthropic-ai/claude-codeThere are also apt, dnf and apk repositories for Debian, Fedora, RHEL and Alpine (in the official installation docs). The npm package downloads the same native binary; to update, use npm install -g @anthropic-ai/claude-code@latest.
| Requirement | Details |
|---|---|
| Operating system | macOS 13+, Windows 10 (1809)+ or Windows Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+ |
| Hardware | At least 4 GB of RAM; x64 or ARM64 processor |
| Network and country | An internet connection; being in a country Anthropic supports (Turkey is on the list) |
| Shell | Bash, Zsh, PowerShell or CMD |
| Account | Pro, Max, Team, Enterprise or Console; not the Free Claude plan |
According to the official release notes, the latest version is 2.1.300 (October 14, 2026). The stable channel is on 2.1.292. Versions change quickly: check yours with claude --version and the update status with claude doctor; to update manually, type claude update.
The native installer updates itself. You change the update channel with the "Auto-update channel" option in /config (latest or stable). A newly released model sometimes requires a newer version than the stable channel has.
Installation takes one command, but it runs the program you download. Only copy the command from the official docs (claude.ai/install.sh, claude.ai/install.ps1); don't run a command with a different address that you've seen on another site.
Your First Command and Checking the Installation
- Check that it's installed: open a new terminal and type
claude --version. If it works, it shows the version number followed by "(Claude Code)". If you get a "not found" error, the installation directory isn't on your PATH (see "Common problems" below). - Detailed check:
claude doctorreads the installation health and settings without starting a session. - Go to your project folder and start: in the terminal,
cd folder-name, thenclaude. - Log in: on first launch a browser opens; confirm the login with your Claude account. Your credentials are saved, so you don't log in every time. To switch accounts, type
/logininside the session. - Ask your first question: try a question like "what does this project do?" (you can ask in Turkish too). Claude reads your files itself; you don't need to add context piece by piece.
› claude --versionIf you've never opened a terminal, the official "terminal guide" page shows step by step how to open one and paste a command.
How Do You Use Claude Code?
First have it explain the project, then ask for a small task. When a change comes in, Claude shows it; if it asks for approval, say Yes. If the mode means it doesn't ask, you can still see what it did afterward with /diff. You can change the session's permission mode at any time with Shift+Tab.
- Be specific: instead of "fix the bug," say "a blank screen appears when the wrong password is entered; fix that."
- Ask step by step: number a big task as 1., 2., 3.
- Let it explore first: before any changes, have it understand the project with a request like "analyze the database schema."
- Put lasting information in a file: write the project's rules in a
CLAUDE.mdfile; Claude reads it every session./initproduces a first draft.
!Claude Code can make mistakes. On an important project, track changes with Git and read them with /diff before committing; /rewind is there to undo.
More ready-made prompts: image and photo prompt examples
Claude Code Commands: /clear, /compact, /model and More
There are two kinds of commands. What you type in the terminal to start with claude are shell commands; what you type with / while Claude Code is open are slash commands. The ones below are selected from the official command list; for the full list, type /help in a session.
| Shell command | What it does |
|---|---|
claude | Starts an interactive session |
claude "task" | Starts a session with an initial prompt |
claude -p "query" | Asks a single question, prints the answer and exits |
claude -c | Continues the most recent conversation in this folder |
claude -r | Lets you pick an earlier conversation and resume it |
claude update | Updates Claude Code |
claude doctor | Checks the installation and settings without starting a session |
claude mcp add | Adds an MCP server (see below) |
| Slash command | What it does |
|---|---|
/help | Shows the commands |
/clear | Starts a new conversation with empty context; use it when switching to a different task |
/compact | Frees up context by summarizing the conversation; you keep going in the same conversation |
/context | Shows what's filling the context as a colored grid |
/init | Starts a CLAUDE.md guide for the project |
/model | Changes the model and makes it the default for future sessions |
/effort | Sets the effort (thinking) level |
/plan | Switches to plan mode: plan first, then approval |
/permissions | Manages permission rules (allow / ask / deny) |
/hooks | Shows the configured hooks |
/mcp | Manages MCP server connections |
/memory | Edits CLAUDE.md files and auto memory |
/config | Opens theme, model and other preferences |
/usage | Shows session cost, plan limits and activity stats (/cost is the same) |
/status | Opens version, model, account and connection status |
/resume | Resumes an earlier conversation |
/rewind | Rolls the conversation and/or code back to an earlier point (/undo is the same) |
/diff | Shows changes in the working tree, including Claude's |
/review | Reviews the current changes for bugs |
/security-review | Reviews the changes on your branch for security vulnerabilities |
/add-dir | Adds another working folder to the session |
/voice | Turns on voice dictation; requires a Claude.ai account |
/desktop | Continues the session in the desktop app |
/doctor | Diagnoses installation, settings and CLAUDE.md problems |
/login, /logout | Logs in to or out of your account |
/release-notes | Shows the release notes |
/exit | Exits Claude Code |
Source: code.claude.com/docs/en/commands (October 14, 2026). The list changes between versions; some commands may be removed in newer versions. Some commands depend on your plan or version (e.g., /privacy-settings is Pro and Max only).
Which Model Does Claude Code Use? Fable, Opus, Sonnet and Haiku
Claude Code uses Claude's models. Type /model in a session and pick from the list; you can also set it at launch, like claude --model sonnet. The default model on Pro, Max, Team, Enterprise and API accounts is Opus 5.5. A bigger model thinks more deeply but is slower and uses up your allowance faster.
| Model (alias) | What it's for, according to Anthropic | Speed | Which plan |
|---|---|---|---|
Fable 5.1 (fable) | The hardest, longest-running work; tasks too big to fit in one session | Slowest | On Max, up to half of your weekly allowance · on Pro, only with extra usage credits |
Opus 5.5 (opus) | Complex reasoning, long-running coding work; Claude Code's default | Medium | Pro and up |
Sonnet 5.5 (sonnet) | Everyday coding; a balance of speed and intelligence | Fast | On paid plans |
Haiku 5.5 (haiku) | Simple, high-volume tasks | Fastest | On paid plans |
- Most coding work, everyday changes: Sonnet
- Complex design decisions, multi-step work: Opus
- Very large, long-running work (on Max): Fable
- Simple, repetitive work: Haiku
- Plan with Opus first, then implement with Sonnet:
opusplan
In Claude Code, Opus uses noticeably more of your allowance. Fable uses it up faster than the others. The effort level also has an effect.
/model opens the list; Enter selects the model and makes it the default, s selects it for this session only. You can type /model sonnet directly. You can also launch with claude --model opus.
With /effort you choose low, medium, high, xhigh or max. Opus 5.5, Sonnet 5.5 and Haiku 5.5 start at medium by default. max applies to that session only. A higher level is more thorough but uses more of your allowance.
New models require a newer Claude Code version: v2.1.257+ for Fable 5.1, v2.1.280+ for Opus 5.5, v2.1.284+ for Sonnet 5.5, v2.1.293+ for Haiku 5.5. If you can't select a model, type claude update.
Source: code.claude.com/docs/en/model-config (October 14, 2026) and the Claude models overview (speeds and descriptions, October 13, 2026). Model versions change often; this section was checked on October 14, 2026.
Can Claude Code Delete My Files? Permissions, Plan Mode, MCP and Hooks
Claude Code edits files and runs commands, which means the responsibility sits on your computer. The permission mode decides when it asks. The mode is chosen per session and you can change it at any time with Shift+Tab.
!In v2.1.283 and later, terminal and VS Code sessions open in auto mode by default: Claude works without asking for approval, and a second model checks its actions in the background. The docs say this "does not guarantee safety." If you want to see every step, switch to Manual mode with Shift+Tab.
| Mode | What it does without asking | When |
|---|---|---|
| Manual | Reading only | Moving carefully, sensitive work |
| Accept edits | Reading, editing files, commands like mkdir/mv/cp | Repetitive work in code you're reviewing |
| Plan | Reading; no changes | Understanding the project and drafting a plan first |
| Auto | Everything; a second model checks in the background | Long tasks, fewer approval prompts |
| Don't ask | Only pre-approved tools | CI running a fixed list of commands |
| Bypass permissions | Everything, unchecked | Only in an isolated container or virtual machine |
- In Manual mode, reading and writing outside your working folder requires approval first
- Network commands like
curlandwgetaren't auto-approved - The first time you open an unfamiliar folder, a "trust" prompt appears
- The sandbox that restricts Bash commands (
/sandbox) is available on macOS, Linux and WSL 2, not on native Windows
With /permissions you write rules for which tools are allowed and which are blocked. "Deny" rules apply in every mode.
MCP connects Claude Code to external tools (issue trackers, databases, design tools). To add one: claude mcp add --transport http <name> <url>. Servers in a project's .mcp.json require separate approval. Anthropic reviews the connectors in its directory against its criteria but doesn't perform security audits: don't connect a server you don't trust.
A hook is a setting that automatically runs your own command at specific moments (e.g., when a file changes); it's written in settings.json and listed with /hooks. The command runs on your computer: read any hook you got from someone else before adding it.
settings.json; it isn't a standalone file.Is Claude Code Free? Plans and Pricing
No. Claude Code isn't on the Free Claude plan. It works in one of two ways: a paid Claude subscription (Pro, Max, Team, Enterprise) or a pay-as-you-go Console account. With a subscription, usage comes out of the same pool as the chat app.
- Chat app
- Claude Code
- Claude Code (terminal, IDE, desktop, web)
- Opus
- 5-hour + weekly limit
- Fable only with extra credits
- 5x the usage of Pro
- Fable (up to half of the weekly allowance)
- $100 a month in API credits (can't be used in Claude Code)
- Prepaid credits
- Per-token pricing
- Billed separately from the subscription
View as a table (including Max 20x and annual pricing)
| Plan | Web (excl. tax) | In-app purchase | Claude Code |
|---|---|---|---|
| Free | $0 | Free | No |
| Pro | $20/month ($17/month billed annually) | Available (monthly and annual) | Yes |
| Max 5x | $100/month | Available (monthly) | Yes |
| Max 20x | $200/month | Available (monthly) | Yes |
| Team, Enterprise | On the Claude pricing page | Not in the store | Yes (on Enterprise, depending on seat type) |
In-app prices come only from the App Store (if you subscribe inside the app, Apple bills you). If you buy on the web, the price is in US dollars and excludes tax; tax is added based on your card's billing address. The Claude guide has the details on the amount and tax.
Claude guide: plans, tax and App Store pricing
The official pages don't list a separate student program or discount code for Claude Code. The Claude guide also notes that there's no free or discounted offer for individual students in Turkey.
Prices from claude.com/pricing and the App Store listing, October 14, 2026. Prices and plan contents may change; check the current page before you pay. This is not financial advice.
Official pricing page last checked on .
What Happens When You Hit the Claude Code Limit? How Does API Pricing Work?
Pro and Max have a 5-hour session limit and a weekly limit. Max also has a separate weekly limit for Fable. Your chats in the Claude app and Claude Code draw from the same pool. There's no fixed number of messages.
Type /usage in a session: it shows your plan limits and what's using up your allowance. You get a warning as you approach the limit.
If you hit the five-hour limit in the middle of a response, Claude Code may show "Usage limit reached · wrapping up" and keep working briefly to finish the task. This allowance is small and limited, and it comes out of your weekly limit; it doesn't apply to new messages.
- Wait for the window to reset
- Upgrade your plan (Pro → Max 5x → Max 20x)
- Turn on usage credits: you continue at standard API prices (
/usage-credits,/rate-limit-options) - Switch to a Console account and buy API credits
!If the ANTHROPIC_API_KEY environment variable is set on your computer, Claude Code uses that key instead of your subscription and charges API fees. To have it use your subscription, remove the variable, type /logout and log in again with your Claude account.
Through Console or credits, Claude Code charges per token processed (a chunk of text the AI reads and writes). Costs vary widely with the model you choose, the size of the project and your usage. The dollar amount in /usage is an estimate; for the exact bill, check the Usage page in Console.
Anthropic states that in enterprise deployments the average is about $13 per developer per active day and $150–250 per month, with 90% of users staying under $30 a day. This is a company average, not a guarantee of your bill.
The monthly API credits that come with the Max plan can't be used in Claude Code or the Claude apps.
Reported by users People using it through the API write that the bill grows quickly, with some spending tens or even hundreds of dollars on a single task; one user later switched to a Max subscription.
- Type
/clearwhen the topic changes. Context left over from the old topic is reprocessed with every new message and uses up your allowance. If you want to come back to it, first/rename, then/resume. - If the same topic runs long, type
/compact. It summarizes the conversation. You can say what to keep:/compact keep the code examples. - Pick a lighter model for simple work. The docs say most coding work is done well with Sonnet and recommend saving Opus for complex work (
/model). - Turn off MCP servers you don't use. See them with
/mcp, and check what's filling the context with/context.
Does Claude Code Work in Turkish?
| Replies in TurkishYes | If you write in Turkish, you get replies in Turkish. If you want it to speak Turkish in every session, enter the language name in the language setting in /config or settings.json; the docs say the value isn't validated and any language name Claude can read works. Code, commands and file names stay in English. |
|---|---|
| Interface language (menus)Unclear | Whether the terminal interface has a Turkish option is not stated in the official documentation; the docs are in English. No setting that switches the menus to Turkish appears in the docs. |
| Voice dictationYes | Dictate by speaking with /voice. Turkish (tr) is among the supported languages. The language setting determines the language; without it, dictation transcribes in English. Requires a Claude.ai account. |
Commands (/clear, /model) and error messages are in English. The tables on this page explain what they do.
Where Does My Code Go? Data, Training Settings, Commercial Use
While it works, Claude Code sends your prompts, Claude's responses and the relevant code snippets to Anthropic over the internet (encrypted with TLS). Your project files stay on your computer, but the parts Claude reads are sent to produce a response.
On these accounts, if the "improve the model" setting is on, your Claude Code sessions may also be used to train new models. You change the setting at claude.ai/settings/data-privacy-controls or with /privacy-settings in a session (Pro and Max only).
Under commercial terms, Anthropic doesn't train models on code and prompts sent to Claude Code (unless you opt into a sharing program yourself).
On consumer accounts, 5 years if you allow training, 30 days if you don't. On Team, Enterprise and API, the standard is 30 days. Zero data retention (ZDR) is only for eligible enterprise accounts and is enabled separately.
Session transcripts are kept as plain text under ~/.claude/projects/ for 30 days (the period is configurable). If a session contains sensitive information, consider clearing that folder too.
Conversations and code you send with /feedback, /bug and /share go to Anthropic and are kept for 5 years. If you answer Yes to the session survey's question about whether your transcript may be reviewed, it's uploaded; if you answer No or Don't ask again, nothing is sent.
Usage telemetry is on by default but contains no code, prompts or file paths; it's turned off with DISABLE_TELEMETRY=1. With Pro and Max logins, error reporting is on too (turned off with DISABLE_ERROR_REPORTING=1); patterns that look like secrets are stripped before sending.
- Using the code you have Claude Code write in your business (the Consumer Terms assign to you their rights in the output, if any)
- Building your commercial product or a client's site
- Using Claude Code to build a competing AI model or product
- Reselling the service or the account
- Sharing the account with more than one person
The software is closed source: the license file in the repository says "all rights reserved" and ties its use to Anthropic's Commercial Terms. You're responsible for the accuracy of the output and for making sure it doesn't infringe anyone else's rights.
Sources: code.claude.com/docs/en/data-usage, security (October 14, 2026); for output rights and prohibitions, Consumer Terms Sections 3–4 (October 13, 2026, Claude guide). This is not legal advice. If you'll work with client data regularly, look at the commercial terms for Team, Enterprise or the API.
Common Problems and Fixes
| What's happening | What to do |
|---|---|
claude command not found | Open a new terminal window. If it's still not found, the installation directory isn't on your PATH; see the PATH section of the official "Troubleshoot installation" page. |
&& error in PowerShell | You ran the CMD command in PowerShell. For PowerShell, use irm https://claude.ai/install.ps1 | iex. |
irm not recognized in CMD | You ran the PowerShell command in CMD. For CMD, use the curl ... install.cmd command. |
The install command returns 403 or syntax error | The official "Troubleshoot installation" page has a list of fixes by error message and alternative installation methods. |
| I have a subscription but I'm being charged API fees | The ANTHROPIC_API_KEY variable is set. Remove it, type /logout, run claude again and log in with your Claude account. |
| The option to log in with my subscription doesn't appear | Log out (/logout), run claude update, close and reopen the terminal completely, and type claude. |
| My allowance runs out quickly | Check with /usage; use /clear when the topic changes, /compact in long conversations, and a lighter model for simple work. |
| Model not in the list / "does not support this model" | New models require a newer Claude Code version. Type claude update. |
| Unexpectedly high API bill | Check the Usage page in Console. /usage is an estimate. Check whether an API key is being used instead of your subscription. |
| My Homebrew or WinGet installation is out of date | These methods don't update themselves: run brew upgrade claude-code or winget upgrade Anthropic.ClaudeCode. |
How Do You Cancel and Uninstall Claude Code?
Claude Code isn't a separate subscription; it's part of your Claude plan. What you cancel is the Pro or Max plan; you uninstall the program separately.
- Cancel the subscription: on claude.ai, click your name in the bottom left › Settings then Billing then Cancel. If you subscribed through an app store, cancel in your Apple or Google subscription settings.
- Mind the timing: cancellation takes effect at the end of the billing period; to avoid a new charge, cancel at least 24 hours before your next billing date.
- Uninstall the program: run the command below that matches how you installed it.
› rm -f ~/.local/bin/claude
rm -rf ~/.local/share/claude› Remove-Item -Path "$env:USERPROFILE\.local\bin\claude.exe" -Force
Remove-Item -Path "$env:USERPROFILE\.local\share\claude" -Recurse -Force› brew uninstall --cask claude-code› winget uninstall Anthropic.ClaudeCode› npm uninstall -g @anthropic-ai/claude-codeAnthropic's policy: payments are generally non-refundable; you submit a request through Get help › "Claude Refund Request." If you bought through the App Store, only Apple can issue a refund. The monthly API credits that come with Max aren't refundable and expire at the end of the period. A separate refund rule for Turkey is not stated in the official documentation.
Uninstalling the program doesn't delete the settings and past conversations on your computer (the ~/.claude folder); delete them manually if you want.
Claude Code, Cursor or OpenCode?
These are tools that have AI write code, or that run somewhere other than Claude. The comparison below draws only on the guides on this site; there's no guide or official source on this site yet for a comparison with OpenAI Codex, so prices and limits aren't compared.
Frequently asked questions
How do you install Claude Code?
On macOS, Linux and WSL, type "curl -fsSL https://claude.ai/install.sh | bash"; in Windows PowerShell, type "irm https://claude.ai/install.ps1 | iex"; then check it in a new terminal with "claude --version". There are also Homebrew ("brew install --cask claude-code") and WinGet ("winget install Anthropic.ClaudeCode") options.
What is Claude Code and what is it for?
It's Anthropic's coding tool. It reads and changes the files in your project folder and runs commands and tests. It works in the terminal, in VS Code and JetBrains, in the desktop app and in the browser.
How do you use Claude Code?
Type "claude" in a terminal in your project folder, log in through the browser, then write what you want in a plain sentence. When a change comes in and it asks for approval, say Yes. Inside the session, /help shows the commands and /clear starts a new topic.
Is Claude Code free?
No. It isn't on the Free Claude plan; you need Pro, Max, Team, Enterprise or a pay-as-you-go Console (API) account.
How much does Claude Code cost?
It has no separate price; it's included in Claude plans. On the web, Pro is $20 a month plus tax, Max 5x is $100 and Max 20x is $200 a month. If you use it through the API, you pay per token.
Which model does Claude Code use?
On Pro, Max, Team, Enterprise and API accounts, the default is Opus 5.5. By typing /model you can choose Sonnet, Haiku or (depending on your plan) Fable.
Does Claude Code work in Turkish?
If you write in Turkish, you get replies in Turkish; to make it permanent, enter the language name in the language field in the settings. Whether the terminal interface has a Turkish option is not stated in the official documentation. Voice dictation supports Turkish.
What happens when you hit the Claude Code limit?
Pro and Max have a 5-hour and a weekly limit; chat and Claude Code draw from the same pool. When you hit it, you wait, upgrade your plan, turn on usage credits or switch to a Console account. You can see how much is left with /usage.
Can Claude Code delete my files?
Technically yes, since it can run commands. In the default auto mode a second model checks its actions, but that doesn't guarantee safety. To see every step, switch to Manual mode with Shift+Tab, use Git on important projects, and undo with /rewind.
I have a subscription, so why am I being charged API fees?
If the ANTHROPIC_API_KEY environment variable is set on your computer, Claude Code uses that key instead of your subscription. Remove the variable, type /logout and log in again with your Claude account.
Does Claude Code work on Windows?
Yes. On Windows 10 (1809) and later it installs natively with PowerShell or CMD; WSL isn't required. The only thing missing on native Windows is the sandbox, which is available in WSL 2.
Is Claude Code free for students?
The official pages don't list a student program or discount code for Claude Code; according to the Claude guide, there's no free offer for individual students in Turkey. If your university has a Claude for Education agreement, check your school account.
Tasks and setup kits that use Claude Code
Related guides
Sources
- Claude Code installation (Advanced setup)
- Claude Code quickstart
- Claude Code overview
- Commands list
- Permission modes
- Security
- Model configuration
- Cost management
- Data usage
- MCP
- Hooks guide
- Voice dictation
- Claude Code on the web
- Release notes
- Using Claude Code with a Pro or Max plan
- Canceling Pro or Max
- Claude pricing
- Supported countries
- Claude on the App Store (Turkey)
- Claude Code repository and license
- Terms of Use (consumer)
Not clear in the official sources: A Turkish option for the terminal interface (not stated in the official documentation; see the "Does Claude Code Work in Turkish?" section) · Claude Code coverage in Team and Enterprise seat prices (the pricing page only says Yes) · refunds and expiry of prepaid Console credits · a refund rule for Turkey · using Claude Code with a third-party model provider (the docs only cover LLM gateways and cloud providers; for a specific provider, see the relevant guide). Each one is covered in its section, along with what you can do.





