Claude Code Developer Guide FAQ
All 11 answers
What is Claude Code?
Claude Code is Anthropic’s coding tool that runs in your terminal inside a repository. It reads your files, edits them and runs commands such as tests, asking for permission according to your settings. Install it with the native installer, Homebrew or npm (Node.js 22 or later), then run claude in the repo and log in through the browser. It needs a paid Claude plan or a Console account: the free claude.ai plan does not include it.
How is Claude Code different from Cursor, GitHub Copilot, or Aider?
This guide does not compare tools feature by feature, because they change often. What Claude Code does is documented: it runs in your terminal inside the repo, reads and edits files and runs commands within your permission rules, and it supports CLAUDE.md, hooks, MCP servers, skills and subagents. To choose, try each tool on a copy of one Magento repo and judge them on the same checks: does it find the right files, respect your deny rules and pass your tests.
What Claude models does Claude Code use?
It uses Anthropic’s Claude models. Which models you can pick depends on your plan and changes over time, so check the model list in the Claude Code docs rather than a fixed list here. Claude Code can also run through Amazon Bedrock, Google Cloud Agent Platform or Microsoft Foundry if your company buys model access there.
How do I install Claude Code?
Pick one install method:
- Native installer:
curl -fsSL https://claude.ai/install.sh | bashon macOS, Linux or WSL, orirm https://claude.ai/install.ps1 | iexin Windows PowerShell. - Homebrew:
brew install --cask claude-code(Homebrew installs do not update themselves). - npm:
npm install -g @anthropic-ai/claude-code, with Node.js 22 or later and withoutsudo.
Check it with claude --version and claude doctor. Then run claude inside your repo and follow the browser prompts to log in.
What is a CLAUDE.md file and why does every project need one?
CLAUDE.md is a Markdown file that Claude Code loads at the start of every session, from ./CLAUDE.md or ./.claude/CLAUDE.md in the project, plus ~/.claude/CLAUDE.md for your own defaults. Put in it what Claude cannot work out from the code: the commands that build and check the repo, conventions such as plugins over preferences, and rules like never editing vendor/. Run /init for a first draft and keep it under 200 lines. It is context, not enforcement: anything that must never happen belongs in a deny rule or a PreToolUse hook.
Are slash-commands like Cursor’s @-mentions?
They do different jobs. Typing @ and a path pulls a file into one prompt. A slash command runs a saved prompt: in Claude Code these are skills, stored at .claude/skills/<name>/SKILL.md (the older .claude/commands/<name>.md format still works), and $ARGUMENTS is replaced by whatever you type after the command. Use @ for context and a skill for a task you repeat.
What are MCP servers and why should I use them?
MCP (Model Context Protocol) servers connect Claude Code to other tools and data, such as an issue tracker or a database. Add one with claude mcp add, for example claude mcp add --transport http <name> <url>. The default scope is local to you; project scope writes .mcp.json in the repo root so the team shares it, and its values can reference environment variables so secrets stay out of the file. Only add servers you trust: a server that fetches outside content can expose you to prompt injection.
What are hooks in Claude Code?
Hooks are commands that run at fixed points in a Claude Code session, such as PreToolUse (before a tool runs), PostToolUse, Stop or SessionStart. You configure them in .claude/settings.json or your user settings, with a matcher such as Edit|Write. A PreToolUse hook that exits with code 2 blocks the call and its message goes back to Claude, so it can refuse edits under vendor/. A PostToolUse hook can run php -l after each PHP edit. Unlike CLAUDE.md, hooks run every time.
What are sub-agents and when should I spawn them?
Subagents are Markdown files with YAML frontmatter in .claude/agents/ (project) or ~/.claude/agents/ (your own). Each one runs in its own context window and returns a summary, so a long search or review does not fill your main session. The tools field limits what it can use: a reviewer with tools: Read, Grep, Glob cannot edit files or run git diff, so pass it the list of changed files from the main session.
How much does Claude Code cost, Claude Pro vs Team vs API?
Claude Code needs a paid Claude plan (such as Pro, Max, Team or Enterprise) or a Console account billed for API use; the free claude.ai plan does not include it. Prices and usage limits change, so check Anthropic’s current pricing before you choose. Teams that already buy model access through Amazon Bedrock, Google Cloud Agent Platform or Microsoft Foundry can use that instead.
What are the best Claude Code workflows for ecommerce dev?
Workflows that suit a Magento repository:
- Ask for a plan before any change that touches several files.
- Give Claude a check to run, such as
php -l, a PHPUnit path orphpcs --standard=Magento2, and ask for the command and its output. - Save a repeated task, such as a new plugin class, as a skill with
$ARGUMENTS. - Review a branch with a read-only subagent.
- Block
app/etc/env.phpwith a deny rule andvendor/edits with a hook.
The learning path on this site walks through each one on your own repo.
Related FAQ topics
All FAQ topics-
Magento Upgrade
Common questions about upgrading Adobe Commerce / Magento Open Source. -
Hyvä Development
Common questions about Hyvä theme development, migration from Luma, and Hyvä Checkout.
-
Magento Performance
Common questions about Magento performance optimization, Lighthouse... -
Emergency Bug Fixing
Common questions about emergency e-commerce bug fixing across... -
Hiring a Freelance Magento Developer
Questions about Magento project scope, quotes, start dates... -
Magento B2B Development
Common questions about Adobe Commerce B2B, Magento B2B development...