Running AI tools
Getting the tools actually running on your machine — installing CLI coding agents, what MCP is, and connecting AI to your own systems.
Installing CLI coding agents: Claude Code, Codex, Gemini CLI, and Kimi
The terminal coding agents worth knowing, what each needs to run (Node version, account), the exact install command for each, and how to pick. With the first-run steps so you get from install to a working agent in a few minutes.
What is MCP, and when do you actually need it?
The Model Context Protocol in plain terms — the USB-C port that lets an AI agent talk to your tools and data. What problem it solves, when it earns its setup cost, and when a plain prompt or a file is enough.
Setting up your first MCP server
A hands-on walkthrough using Claude Code: adding a server with claude mcp add, the difference between local (stdio) and remote (HTTP) servers, choosing a scope, handling auth, and verifying it works.