Installing and Using Claude Code CLI

Claude Code is Anthropic's terminal-native coding agent. Let's install it, wire it to a project, and use the features most engineers miss on day one.

50 min · Reviewed 2026

A Coding Agent That Lives in Your Shell

Claude Code is a terminal-based agent that reads and writes files, runs commands, and carries project context across sessions. Unlike an IDE plugin, it can drive your entire development environment, not just the editor.

Install in under two minutes

# Install globally via npm
npm install -g @anthropic-ai/claude-code

# First run — authenticates via browser
cd my-project
claude

# Inside the TUI, type your first request:
#   "Explain what this repo does and list the main entry points."Two commands and you are running an agent against your codebase.

CLAUDE.md: project memory

Drop a CLAUDE.md file at the root of your project. Claude Code reads it on every session. This is where you put your project's conventions, scripts to run, architecture notes, and the AI etiquette you want followed. Teams treat it as a living style guide.

# CLAUDE.md

## Stack
- Next.js 15, TypeScript, Tailwind, Postgres via Prisma.

## How to run
- `pnpm dev` for local, `pnpm test` for tests, `pnpm build` before PR.

## Conventions
- Server actions live in `app/actions/`, named `<verb><Noun>`.
- Never commit directly to main — always open a branch.
- Use Zod for all public API inputs.

## Things to avoid
- Do not add new dependencies without asking first.
- Do not rename exported functions — downstream consumers break.A CLAUDE.md is the single highest-leverage file in any Claude Code project.

Slash commands and skills

As of 2026, slash commands and skills are unified in Claude Code. Put markdown files in .claude/skills/ at the project root, and each one becomes a /slash-command. The skill defines a repeatable workflow: run tests, deploy, generate a PR description, whatever you need.

# .claude/skills/pr-ready.md
---
name: pr-ready
description: Prepare the branch for PR review
---

1. Run `pnpm lint --fix` and commit any fixes.
2. Run `pnpm test` and make sure all pass.
3. Run `pnpm build` and fix any type errors.
4. Write a PR description summarizing the diff in 3 bullets.
5. Suggest a PR title under 70 characters.Now typing /pr-ready in any Claude Code session triggers this playbook.

Sub-agents: up to 10 in parallel

A sub-agent is a separate Claude instance spawned from your main session to handle a narrow job. Define one in .claude/agents/explore.md with its own tool access and system prompt. Then say: Use the explore agent to find all auth-related files. Up to 10 can run in parallel, which is a game-changer for parallel codebase exploration.

Day-one workflow

Install, authenticate, cd into your project
Write a CLAUDE.md describing stack, conventions, and commands
Ask Claude to explain the repo architecture out loud
Create one skill for a workflow you repeat (tests, deploy, etc.)
Ask Claude to plan a feature, then execute step by step

The best interface is the one that can drive the whole machine. The terminal still is.
— A Unix-pilled engineer

The big idea: Claude Code treats your terminal as the agent's workspace. CLAUDE.md gives it context, skills automate recurring work, and sub-agents run parallel tasks. Invest in those three and the tool compounds.

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-coding-installing-claude-code-creators

What file should you place at your project's root to give Claude Code persistent context about your project's conventions, scripts, and architecture?
1. context.json
2. settings.yaml
3. CLAUDE.md
4. project.config
What is the maximum number of sub-agents that Claude Code can run simultaneously in a single session?
1. 10
2. Unlimited
3. 5
4. 25
Where should you define a custom sub-agent with its own tool access and system prompt?
1. agents.json
2. ~/.claude/default-agent/
3. .claude/agents/explore.md
4. /etc/claude/agents/
What is the primary purpose of the hook system in Claude Code?
1. To connect to external databases
2. To generate code automatically
3. To manage user authentication
4. To trigger custom scripts when specific lifecycle events occur
Which of the following is NOT a lifecycle event supported by Claude Code's hook system?
1. PostCompact
2. FileCompressed
3. SubagentStart
4. FileChanged
What makes sub-agents especially powerful for codebase exploration?
1. They can run in parallel with different permissions
2. They share memory with the main agent
3. They require manual supervision
4. They can only read files, not execute commands
What happens when you place a markdown file in your project's .claude/skills/ directory?
1. It gets converted to Python code
2. It becomes a callable slash command
3. It creates a new sub-agent
4. It is ignored by Claude Code
A team wants to ensure that tests run automatically every time a file is edited in their project. Which feature should they configure?
1. A CLAUDE.md preference
2. A skill in the skills directory
3. A new sub-agent
4. The hook system with a FileChanged event
What is the recommended first step after installing and authenticating Claude Code in a new project?
1. Delete the node_modules folder
2. Configure the hook system
3. Create a CLAUDE.md file describing the project's stack and conventions
4. Start writing code immediately
What type of information should you include in a CLAUDE.md file?
1. Database connection strings
2. Project conventions, scripts to run, architecture notes, and AI etiquette
3. Only function names in your code
4. A list of all files in the project
What distinguishes a sub-agent from the main Claude Code session?
1. A sub-agent runs faster than the main session
2. A sub-agent cannot read files
3. A sub-agent has its own tool access and system prompt definition
4. A sub-agent requires internet access
Which directory structure is correct for defining a skill that can be invoked with a slash command?
1. ~/.claude/
2. /commands/
3. .claude/skills/
4. /skills/
What is the 'big idea' behind Claude Code's design philosophy, as described in the material?
1. Using Claude Code exclusively for deployment tasks
2. Treating the terminal as the agent's workspace, with CLAUDE.md for context, skills for automation, and sub-agents for parallel tasks
3. Making Claude Code act as a replacement for version control
4. Replacing all IDEs with terminal-only workflows
You want one sub-agent to have database access and another to only read repository files. What feature enables this?
1. Scoped permissions through separate MCP server configurations
2. Team plan features
3. Different subscription tiers
4. A setting in CLAUDE.md
Which of these is a recommended day-one workflow for getting started with Claude Code in a project?
1. Install, configure the database, write tests first
2. Install, delete all files, ask Claude to rebuild, deploy
3. Install, create 10 sub-agents, run them all
4. Install, authenticate, create CLAUDE.md, ask Claude to explain architecture, create a skill, plan a feature

← Back to interactive lesson

Tendril · Creators · AI-Assisted Coding