Tutorials
Step-by-step setups for the tools people actually use — cloud agents like Claude Code and Codex CLI, and local stacks like Ollama and LM Studio.
Reading is nice. Running is better.
These tutorials take you from an empty machine to a working setup, with the exact commands and the errors you’ll actually hit.
☁️ Cloud tools & agents
Set up the big cloud players: Claude Code, Codex CLI, Gemini CLI, Copilot, Cursor, Windsurf.
Claude Code: Anthropic's coding agent in your terminal
Install Claude Code with the native installer, sign in with a Claude plan, run your first agentic session, and learn the handful of slash commands that matter. Permissions, CLAUDE.md and headless mode included.
Codex CLI: OpenAI's terminal agent, with a sandbox and a local-model escape hatch
Install Codex CLI, sign in with your ChatGPT plan, learn the sandbox and approval flags that keep it safe, wire it to LM Studio or Ollama with --oss, and run it headless with codex exec.
Gemini CLI: the agent with the free tier that actually works
Install Gemini CLI, log in with a personal Google account, get 1,000 requests a day for free, and learn the @file, !shell and /memory moves that make it a daily driver.
GitHub Copilot: the agent inside the editor you already use
Install Copilot in VS Code, sign in with a GitHub account (free plan included), switch Chat into agent mode, and use the Copilot CLI for terminal work. Plans and AI credits explained without the marketing.
Cursor: the AI-first editor, now sold through The X Subscription
Install Cursor, sign in, run your first Agent task with Cmd/Ctrl+I, and understand the 2026 twist: Cursor is now part of The X Subscription, with new plans from $0 to $200. Legacy Cursor plans still work.
Windsurf is now Devin Desktop: install it and run your first agent
On June 2, 2026, Windsurf became Devin Desktop. Same editor, same extensions, new owner (Cognition). Here's how to install it, import your VS Code or Cursor setup, and use the agent — plus the Free tier's surprising limits.
🏠 Local AI
Run your own models at home: Ollama, LM Studio, Open WebUI, llama.cpp — no cloud, no API key, your data stays yours.
Ollama: run LLMs locally in 3 commands
Install Ollama, pull a model, and chat — from the terminal and from any app, over a local API. Tested on a real Windows machine.
llama.cpp: the engine that runs local AI
Download the official Windows build, load a GGUF model, serve it over the OpenAI API — and measure the speed. Tested on a real CPU-only machine.
Open WebUI: a ChatGPT-style interface for your local models
Install Open WebUI without Docker (the uv path), create your account, and chat with the models you already pulled in Ollama. Tested end-to-end on a real machine.
LM Studio: the full local AI workstation — hub, chat, server, CLI
Install LM Studio, download a model from the Hub, chat in the app, and serve it with the local OpenAI-compatible API on port 1234. Plus the lms CLI that does it all from the terminal.
Hermes Agent: a local AI agent that actually works on your machine
Install Hermes Agent, point it at the local model from the previous tutorials, and give it real jobs — files, terminal, web — with an agent loop you can watch and control.