THE FIELD GUIDES

AI on a Mac

Open Terminal (press Cmd + Space, type Terminal) and paste one line at a time. Jump to: AI on a Mac · AI cloud · AI offline

AI on a Mac

Set up your Mac once: Homebrew, Git, Node, Claude Code, and GitHub and Vercel to publish what you build.

Set up your Mac

STEP 1
Install Homebrew
Homebrew installs everything else with one line each:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
When it finishes, it prints two "Next steps" lines. Copy and run those, then close and reopen Terminal.
STEP 2
Git, Node and Bun
brew install git node
curl -fsSL https://bun.sh/install | bash
Git saves and shares code, Node runs most AI coding tools, and Bun is the fast JavaScript runtime LifeOS needs (step 4).
STEP 3
Claude Code
npm install -g @anthropic-ai/claude-code
Go to your project folder and start it. It opens the browser to sign in with your Claude account:
claude
STEP 4
LifeOS: give Claude Code a memory and your goals
LifeOS (free, by Daniel Miessler) runs on top of Claude Code. It remembers past sessions, keeps your goals in a TELOS file, and turns mistakes into rules. Back up your Claude settings first:
cp -r ~/.claude ~/.claude-backup
Then install it with the one-line installer:
curl -fsSL https://ourlifeos.ai/install.sh | bash
Or start claude and paste this, and Claude does the install for you:
Read https://ourlifeos.ai/install and install LifeOS for me.
Fill in your TELOS (mission, goals, projects) when it asks. The files are at github.com/danielmiessler/LifeOS (green Code button, then Download ZIP).

GitHub and Vercel

STEP 1
Sign in to both
brew install gh
gh auth login
npm install -g vercel
vercel login

AI cloud

AI that runs on the companies' servers. You sign in or use an API key, and you need internet.

The quickest way: chat in your browser

STEP 1
Open ChathoNU
Go to chathonu.com to chat with Wednesday, FRIDAY and the rest of the crew for free. No install needed.

CLI AI coding tools

Every AI coding tool you can run in Terminal. Works on every Apple Silicon Mac (M1 to M5). Do steps 1 and 2 of "Set up your Mac" first. Install, then start it with the second line. Each asks you to sign in or paste an API key the first time.

STEP 1
First: install uv (for Kimi and Mistral)
brew install uv
If kimi or vibe isn't found after installing, run this and reopen Terminal:
uv tool update-shell
STEP 2
Claude Code (Anthropic)
npm install -g @anthropic-ai/claude-code
claude
STEP 3
Codex (OpenAI)
npm install -g @openai/codex
codex
STEP 4
Gemini CLI (Google)
npm install -g @google/gemini-cli
gemini
STEP 5
GitHub Copilot CLI (GitHub / Microsoft)
npm install -g @github/copilot
copilot
Needs a Copilot subscription. Then type /login.
STEP 6
Grok Build (xAI)
curl -fsSL https://x.ai/cli/install.sh | bash
grok
STEP 7
Kimi Code CLI (Moonshot)
uv tool install kimi-cli
kimi
Needs uv (the first step in this list). Then type /login and paste your Kimi API key.
STEP 8
Qwen Code (Alibaba)
npm install -g @qwen-code/qwen-code@latest
qwen
Then type /auth. It needs an API key.
STEP 9
Mistral Vibe (Mistral)
uv tool install mistral-vibe
vibe
Needs uv (the first step in this list).
STEP 10
OpenCode (works with any provider)
npm install -g opencode-ai
opencode
Then type /connect to sign in to a provider.
STEP 11
Deep Code (DeepSeek, community tool)
npm install -g @vegamo/deepcode-cli
deepcode
DeepSeek has no official CLI; this community one is listed in DeepSeek's API docs. It needs a DeepSeek API key.

AI offline

Free AI models that run on your Mac itself. Nothing leaves your computer, and it works without internet.

The quickest way: in your browser

STEP 1
Open the offline chat
Go to chathonu.com/local, pick the model marked BEST PICK, tap DOWNLOAD, then USE. It runs on your Mac and keeps working with Wi-Fi off.

Ollama: free models on your own Mac

STEP 1
Install and run a model
brew install --cask ollama
Open Ollama from Applications once, then:
ollama run gemma3:4b
Or run a coding tool on it (the tool must be installed first):
ollama launch claude
STEP 2
Use your Mac's Ollama in ChathoNU
launchctl setenv OLLAMA_ORIGINS "https://chathonu.com"
Quit Ollama from the menu bar and open it again. Then open chathonu.com/local, choose MY OLLAMA and connect to:
http://localhost:11434

Run Claude Code fully offline with a local model

This is how I ran Claude Code on my Mac with no internet, no API key and no subscription in the video. You point Claude Code at a coding model running in Ollama. It's slower than the cloud (a big answer can take a few minutes on-device), but it works in airplane mode.

STEP 1
Pull a coding model
With Ollama installed (above), download a coding model. The 7B is a good balance on an Apple Silicon Mac:
ollama pull qwen2.5-coder:7b
STEP 2
Optional: give it a name and personality with a Modelfile
A Modelfile layers your own system prompt and settings on top of a base model. Make one:
nano Modelfile
Paste this (a friendly, gamer-style coding assistant), then save with Ctrl+O, Enter, Ctrl+X:
FROM qwen2.5-coder:7b # A coding buddy with a Sword-Art-Online gamer personality. SYSTEM """ You are Kirito, a sharp and upbeat coding assistant. Explain code like a game: bugs are monsters, debugging is clearing a dungeon. Always give working, well-commented code and a short, clear explanation. Stay positive and keep the player moving. """ PARAMETER temperature 0.7 PARAMETER num_ctx 8192
Build it into a named model:
ollama create kirito -f Modelfile
Try it on its own:
ollama run kirito

Later, to see this Modelfile again: ollama show kirito --modelfile. To back it up: ollama show kirito --modelfile > ~/Documents/kirito.Modelfile.

STEP 3
Point Claude Code at your local model
Claude Code reads these two settings. Set them so it talks to Ollama instead of the cloud (use kirito, or qwen2.5-coder:7b if you skipped the Modelfile):
export ANTHROPIC_BASE_URL="http://localhost:11434"
export ANTHROPIC_MODEL="kirito"
To make it stick every time, add those two lines to ~/.zshrc, then reopen Terminal.
STEP 4
Turn off the update check (so it doesn't hang offline)
echo 'export DISABLE_AUTOUPDATER=1' >> ~/.zshrc && source ~/.zshrc
The "Auto-update failed" line you see offline is harmless; this just hides it.
STEP 5
Start coding offline
Make sure Ollama is open, turn off Wi-Fi if you want to prove it's local, then in your project folder:
claude
Ask it anything. On-device answers are slower because they run on your Mac, not a server.

Use your Mac's Ollama from your phone

STEP 1
Install Tailscale on the Mac and your phone
brew install --cask tailscale
Sign in to the same account on both.
STEP 2
Give Ollama an https address
tailscale serve --bg 11434
Use the https address it prints in MY OLLAMA on chathonu.com/local.

If the tailscale command isn't found, open the Tailscale app once first.

Stuck on a step? Send a message and say which one.