Lesson 454 of 2116
Installing And Authenticating Claude Code
Setup is short — but the setup choices shape every session afterwards. Get the model, billing, and permissions right on day one.
Lesson map
What this lesson covers
Learning path
The main moves in order
- 1The 5-minute install
- 2installation
- 3authentication
- 4billing model
Concept cluster
Terms to connect while reading
Section 1
The 5-minute install
Claude Code installs as a global CLI. After install you authenticate against your Anthropic account, pick a billing source, and run it inside any project directory. The first run will offer to create a CLAUDE.md and explain how it intends to use your shell.
Install
The minimal install + auth flow. The exact package name and command may shift; the docs page is canonical.
# Install via the official installer (check the latest docs page first)
npm install -g @anthropic-ai/claude-code
# Authenticate
claude login
# Run inside a project
cd ~/code/my-project
claudeTwo billing modes
- 1Subscription: a Claude.ai or Claude Pro / Max plan covers usage with bundled limits — predictable cost, capped throughput
- 2API billing: the CLI uses your Anthropic API key directly, so usage is per-token — no caps but no bundle either
- 3Most individual users start on the subscription path
- 4Teams that hit subscription limits switch to API billing for the heavy hitters
Permissions on first run
On first run, Claude Code asks how aggressively it should ask before running commands. The default is 'ask before any non-read, non-trivial command.' Conservative settings are the right starting point — you can loosen permissions per command type as trust grows.
Verify the install
- Run `claude --version` to confirm the binary is on your PATH
- Run `claude doctor` (if available) to surface common config issues
- Run a one-shot test: ask it to list the files in the current directory
- Confirm the trace shows the read tool firing, not made-up output
Key terms in this lesson
The big idea: pick a billing mode you understand, accept conservative defaults, and verify the trace before trusting any output.
End-of-lesson quiz
Check what stuck
15 questions · Score saves to your progress.
Tutor
Curious about “Installing And Authenticating Claude Code”?
Ask anything about this lesson. I’ll answer using just what you’re reading — short, friendly, grounded.
Progress saved locally in this browser. Sign in to sync across devices.
Related lessons
Keep going
Creators · 10 min
Installing OpenClaw And Wiring It To A Local Model
Get OpenClaw running on your machine in under fifteen minutes, paired with a local LLM via Ollama. The shape of the install matters less than what you verify after.
Creators · 45 min
Structured Outputs: Make the Model Return Data You Can Trust
For production apps, pretty prose is often the wrong output. Learn when to use structured outputs, function calling, and schema validation.
Creators · 9 min
What Perplexity Is: Search-Augmented LLM, Not A Chatbot
Perplexity is built around the idea that every answer should cite its sources. Treating it like ChatGPT misses the point — and the reliability gap that comes with it.