AI Coding Tool Setup and Reference (1/7) — Claude Code Install: macOS, Linux, WSL, Windows

Every install path, system requirement, auth option, and first session — straight from the official docs

ํ•ต์‹ฌ ์š”์•ฝ

  • Audience: developers or non-developers installing Claude Code for the first time.
  • What you'll get: five install methods compared, verbatim per-platform commands, four auth options, and a working first session in under 10 minutes.
  • Prerequisite: a Claude Pro / Max / Team / Enterprise subscription, or an Anthropic Console API account. The free Claude.ai plan does not include Claude Code.

1. Install methods at a glance

The official docs list five install paths. Pick one based on your platform and workflow.

Method Platforms Auto-update When to use
Native Install (recommended) macOS, Linux, WSL, Windows ✅ background Most personal dev setups
Homebrew macOS, Linux ❌ manual When you already manage tools with Homebrew
WinGet Windows ❌ manual When corporate policy requires WinGet
Linux packages (apt/dnf/apk) Debian/Ubuntu, Fedora/RHEL, Alpine ❌ via system upgrade Server automation / deployment pipelines
npm (deprecated) All ❌ manual (npm update) Only if you already have a Node-based workflow

The official README marks npm install as deprecated. Existing npm users should migrate to the native installer.

2. System requirements

Per official docs as of 2026-04-24:

  • OS
  • macOS 13.0+
  • Windows 10 1809+ or Windows Server 2019+
  • Ubuntu 20.04+ / Debian 10+ / Alpine Linux 3.19+
  • Hardware: 4 GB+ RAM, x64 or ARM64
  • Shell: Bash, Zsh, PowerShell, or CMD
  • Native Windows requires Git for Windows
  • WSL does not need Git separately
  • Network: internet connection in an Anthropic-supported country
  • Dependency: ripgrep is bundled in the binary. Alpine and other musl distros need it installed separately (see 2.1)

2.1 Extra step for Alpine / musl distros

Alpine Linux and other musl/uClibc-based distros can't use the bundled ripgrep. Install the system packages instead.

apk add libgcc libstdc++ ripgrep

Then add this to ~/.claude/settings.json:

{
  "env": {
    "USE_BUILTIN_RIPGREP": "0"
  }
}

3. Install — verbatim commands per platform

3.1 macOS / Linux / WSL — Native Install

curl -fsSL https://claude.ai/install.sh | bash

After install, re-open the shell or source ~/.bashrc (or ~/.zshrc) to reload PATH. The default location is ~/.local/bin/claude.

To pin a version:

curl -fsSL https://claude.ai/install.sh | bash -s stable

curl -fsSL https://claude.ai/install.sh | bash -s 2.1.89

3.2 Windows — PowerShell

irm https://claude.ai/install.ps1 | iex

Your prompt shows PS C:\> in PowerShell. If you see 'irm' is not recognized, you're in CMD — use the CMD command instead.

3.3 Windows — CMD

curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd

Your prompt shows C:\> (no PS) in CMD. If you see The token '&&' is not a valid statement separator, you're in PowerShell — switch to the PS command.

Native Windows requires Git for Windows to be installed first. Admin rights are not needed.

3.4 Homebrew (macOS / Linux)

Homebrew offers two casks.

brew install --cask claude-code          # stable (~1 week delayed, skips regressions)
brew install --cask claude-code@latest   # latest (new versions immediately)

Homebrew installations don't auto-update. Run brew upgrade claude-code (or @latest) periodically.

3.5 WinGet (Windows)

winget install Anthropic.ClaudeCode

Upgrade manually: winget upgrade Anthropic.ClaudeCode.

3.6 Linux package managers

Signed repositories are available for Debian/Ubuntu (apt), Fedora/RHEL (dnf), and Alpine (apk).

Debian / Ubuntu (apt)

sudo install -d -m 0755 /etc/apt/keyrings
sudo curl -fsSL https://downloads.claude.ai/keys/claude-code.asc \
  -o /etc/apt/keyrings/claude-code.asc
echo "deb [signed-by=/etc/apt/keyrings/claude-code.asc] https://downloads.claude.ai/claude-code/apt/stable stable main" \
  | sudo tee /etc/apt/sources.list.d/claude-code.list
sudo apt update
sudo apt install claude-code

Verify the GPG fingerprint before trusting:

gpg --show-keys /etc/apt/keyrings/claude-code.asc

Fedora / RHEL (dnf)

sudo tee /etc/yum.repos.d/claude-code.repo <<'EOF'
[claude-code]
name=Claude Code
baseurl=https://downloads.claude.ai/claude-code/rpm/stable
enabled=1
gpgcheck=1
gpgkey=https://downloads.claude.ai/keys/claude-code.asc
EOF
sudo dnf install claude-code

Alpine (apk)

wget -O /etc/apk/keys/claude-code.rsa.pub \
  https://downloads.claude.ai/keys/claude-code.rsa.pub
echo "https://downloads.claude.ai/claude-code/apk/stable" >> /etc/apk/repositories
apk add claude-code

3.7 npm (deprecated)

Requires Node.js 18+. Marked deprecated officially, but still works for existing workflows.

npm install -g @anthropic-ai/claude-code

Avoid sudo npm install -g — it creates permission and security issues.

4. Verify the install

Two commands:

claude --version
claude doctor

claude doctor checks Node version, PATH, permissions, and MCP accessibility. If anything shows up red, see the official troubleshooting page.

5. Auth — four options

Claude Code does not work with the free Claude.ai plan. You need one of:

Option For Billing How to set up
Claude Pro / Max / Team / Enterprise subscription Individuals / teams Flat subscription Run claude → browser login
Anthropic Console (API key) Pay-as-you-go credits Usage-based Console creates a "Claude Code" workspace on first login
Amazon Bedrock Enterprise AWS AWS billing Bedrock setup guide
Google Vertex AI Enterprise GCP GCP billing Vertex setup guide
Microsoft Foundry Enterprise Azure Azure billing Foundry setup guide

On first run, claude prompts you to log in through the browser. To switch accounts later, use /login.

6. First session

cd /path/to/your/project
claude

Run it inside a project directory. The welcome screen shows session info. Memorize these:

  • /help — list available commands
  • /login — switch accounts
  • /clear — reset conversation
  • exit or Ctrl+D — quit

Command cheatsheet

Command Effect
claude Start interactive session
claude "task" Run a one-time task and exit
claude -p "query" One-off query, non-interactive (script-friendly)
claude -c Continue the most recent conversation in the current directory
claude -r Resume a prior conversation by selection

Full reference: official CLI docs.

7. Auto-update & release channels

Native installs update in the background. Two channels:

  • latest (default) — new versions immediately
  • stable — about 1 week delayed, skips releases with major regressions

Switch via /config → Auto-update channel, or edit settings.json directly:

{
  "autoUpdatesChannel": "stable"
}

Disable auto-update entirely:

{
  "env": {
    "DISABLE_AUTOUPDATER": "1"
  }
}

Manual update: claude update.

8. Uninstall

8.1 Native

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

8.2 Full cleanup including config

Deleting ~/.claude and per-project .claude removes all settings, allowed tools, MCP servers, and session history.

rm -rf ~/.claude
rm ~/.claude.json

rm -rf .claude
rm -f .mcp.json

If VS Code extension, JetBrains plugin, or Desktop app is installed, uninstall those first — they recreate ~/.claude when they run.

9. Counter-scenarios — when this guide doesn't fit

  • Corporate firewall blocks the claude.ai install script → Use the Linux package manager path (3.6). Signed repos often pass policy review.
  • You need an offline environment → Claude Code requires an internet connection. There is no fully offline mode in the official docs.
  • You already run a Node-based toolchain → npm install (3.7) still works but is deprecated; plan to migrate to native over time.
  • You're on macOS 12 or older → The minimum is macOS 13.0+. OS upgrade is a hard requirement.

10. What's next

Once installed, these three posts in order:

  1. Claude Code slash commands reference — practical commands beyond /help (coming soon)
  2. CLAUDE.md authoring — what to include, what to avoid — project instruction file design (coming soon)
  3. Claude Code hooks complete guide — 4 lifecycle events × real scripts — automation (coming soon)

References

This is post 1/15 in the "AI Coding CLI Entry Guide" series. Tools covered: Claude Code, Codex CLI, OpenClaw, Hermes Agent. last verified: 2026-04-24 (per official Claude Code docs).

Series overview: Series index

๋Œ“๊ธ€

๋Œ“๊ธ€ ์“ฐ๊ธฐ

์ด ๋ธ”๋กœ๊ทธ์˜ ์ธ๊ธฐ ๊ฒŒ์‹œ๋ฌผ

Agent Memory Engine (2/10) — Building an AI Agent Memory System with SQLite Alone

← 1/10 SQLite Memory Engine f… ๐Ÿ“š Series Index 3/10 memcore Completed → A memory engine that runs without daemons, without dependencies, anywhere ํ•ต์‹ฌ ์š”์•ฝ This post covers how to design an AI agent memory engine (memcore) on a single-file SQLite backend. - Structure: 25 modules, ~6,464 lines, 22 DB tables - Principles: no daemon, single-file portability, no required external dependencies, host-agnostic - Core techniques: 4-layer hybrid search (topic · vector · FTS5 · LIKE), U-tag dialectic-based confidence evolution, ingestion-layer bias rejection rules What This Post Covers One approach to building AI agent memory without a server or cloud vector DB. Specifically: (1) the limitations of text-file-based memory, (2) vector DB alternatives compared and selection criteria, (3) how to layer search tiers, (4) how to evolve confidence scores over time, and (5) how to block biased memory accumulation at the ingestion stage. Design Principles: Why Single-File SQLite Text-file-based ...
Read more »

"ML Foundations (9/9) — PyTorch vs TensorFlow, and the Road to Local LLMs"

์ด๋ฏธ์ง€
← 8/9 Deep Learning Architec… ๐Ÿ“š Series Index (series end) The final part. So far we've focused on models. Now we focus on the tools that actually run them . We'll lay out the philosophical difference between PyTorch and TensorFlow, and trace how Hugging Face transformers , llama.cpp , MLX , and Ollama built the bridge to running large language models on your own machine. By the end you should have the full mental model of "download a pretrained LLM and serve it locally." 0. Learning Objectives Explain the eager-vs-graph difference between PyTorch and TensorFlow. Explain, in graph terms, how autograd automates the backward pass. Use Hugging Face transformers ' AutoModel , AutoTokenizer , and pipeline abstractions. Describe llama.cpp's GGUF quantization, INT4 inference, and CPU-first flow. Describe Apple MLX's use of unified memory and how it differs from PyTorch. Run a local LLM with Ollama and call it through an OpenAI-compatible API. ...
Read more »

"RAG Core Study (14/26) — Evaluation Sets with RAGAS & DeepEval"

์ด๋ฏธ์ง€
Series overview: Series index ๐Ÿ“š Series Index Without an evaluation set, every RAG improvement is a story, not evidence. RAG systems fail in more than one place: retrieval, context selection, answer generation, and grounding. That means "looks better" is not a metric. Part 14 explains how to build a golden dataset , how RAGAS and DeepEval fit into the workflow, and why teams must separate retrieval quality from answer quality if they want tuning decisions to hold up. 0. Prerequisites Part 13 reranking — system quality now depends on multi-stage ranking. Part 1 grounding — a correct-looking answer is not enough. Part 12 Hybrid — multiple retrieval settings need comparison on fixed data. 1. Learning Objectives Build a small but useful golden dataset for RAG. Distinguish retrieval metrics from answer metrics . Use RAGAS and DeepEval in the right roles. Avoid the common traps in judge-model-based evaluation. 2. ํ•ต์‹ฌ ์š”์•ฝ An evaluation set for RAG is ...
Read more »

"ML Foundations (8/9) — Deep Learning Architectures: CNN, RNN, Attention"

์ด๋ฏธ์ง€
← 7/9 Deep Learning Training ๐Ÿ“š Series Index 9/9 PyTorch vs TensorFlow → What did we keep adding on top of an MLP? Vision went CNN, sequences went RNN/LSTM, and eventually both converged on Attention and the Transformer. Each architecture is easier to remember if you read it as what it gave up to gain something else . This part is the one-line summary of the decisive inflection points: LeNet → AlexNet → ResNet → LSTM → Attention → Transformer. 0. Learning Objectives Explain why CNN's convolution, pooling, and weight sharing match images so well. Trace what got unblocked from LeNet → AlexNet → VGG → ResNet as depth grew. State BPTT for RNNs and the vanishing/exploding gradient problem. Write the LSTM gate equations (forget, input, output) with intuition. Write the attention formula in query/key/value form and the scaled dot-product variant. State the precise reasons Transformers replaced RNNs (parallelism + long-range dependencies). 1. ํ•ต์‹ฌ ์š”์•ฝ CNN : weight sharin...
Read more »

"ML Foundations (7/9) — Deep Learning Training: Optimizers, Regularization, Initialization"

← 6/9 Neural Networks ๐Ÿ“š Series Index 8/9 Deep Learning Architec… → Building an MLP, as we did in Part 6, is not the same as getting it to train well . The right optimizer, regularization, initialization, and learning rate — when these line up, deep networks converge. When they don't, the network refuses to learn at all. This part is the catalog of those crafts, with formulas, paper citations, and code in one place. 0. Learning Objectives Compare and write the update rules for SGD, Momentum, Nesterov, and Adam. Explain how Dropout, BatchNorm, and LayerNorm work and where they belong in a model. Derive the variance formulas for Xavier and He initialization and match them to activations. Implement step, cosine, and warmup learning-rate schedules in PyTorch. Explain why gradient clipping is effectively required for RNNs and Transformers. Diagnose the most common training failures (NaN, plateau, overfitting) and apply first-line fixes. 1. ํ•ต์‹ฌ ์š”์•ฝ SGD : \(w \leftarro...
Read more »

OpenClaw to Hermes Migration (2/13) — What to Preserve, Partially Port, or Discard

← 1/13 Current Structure Inve… ๐Ÿ“š Series Index 3/13 What Moves to Hermes → A code review record classifying operational scripts into three tiers: preserve, partial-preserve, and discard. What This Post Covers An asset classification framework (preserve / partial-preserve / discard) for migrating a legacy agent system to a successor A concrete case applying the Strangler Fig + Subset Migration strategy to a memory pipeline Classification results evaluated across LOC, external dependencies, and replaceability axes A method for slimming down a bloated single file (1,310 LOC) using the migration as a natural entry point Problem Definition: Port Everything or Cut Deep? The OpenClaw inventory comprises multiple agents, multiple scheduler/daemon processes, and multiple operational scripts. When moving to the successor system, Hermes, the choice is not binary. A full port carries accumulated technical debt along with it; a minimal port discards hard-won domain logic. Code-revie...
Read more »

AI Agents I Built (5/7) — Building an Automated Blogger API Publishing System

← 4/7 My Life Organizer Agent ๐Ÿ“š Series Index 6/7 Dual → Designing a Markdown-Based Auto-Publish Pipeline with Blogger API v3 + OAuth 2.0 Key Summary Blogger API v3 + OAuth 2.0 enables a pipeline that converts Markdown files into publishable HTML posts blogger_publish.py handles the full flow — frontmatter parsing → body cleanup → HTML conversion → CSS injection → API post — in a single script For bulk publishing, quota management (delay, retry, TOC suppression) is the critical design concern Platform Selection Criteria API access is mandatory for AI agent-driven content publishing. A comparison of platforms: Platform API Cost AdSense Decision WordPress.com REST API Paid plan required Paid only Cost overhead Ghost REST API Self-hosted or paid Manual setup Ops overhead Blogger v3 API Free Native integration Adopted Blogger is free, exposes a REST API, and integrates natively with AdSense. For personal blog automation, this combination i...
Read more »