AI Workshop
Episode 1 of 10Aired

Setup from scratch

A clean machine → a working agent setup in one stream: install, environment, CLAUDE.md and AGENTS.md.

Wednesday 7 October at 19:00 Moscow time· 90 min

Aired

The recording is being prepared — it usually appears within two days. Sign up for the next streams so you don't miss them.

What we'll show

  1. Installing Claude Code and Codex CLI on a clean machine, updating, checking the version
  2. Terminal, editor, git and Node — the minimum an agent can't see without
  3. A first CLAUDE.md: stack, build and test commands, prohibitions — and how it differs from AGENTS.md
  4. How the agent reads a project in its first minutes and why repo hygiene matters
  5. A first small task from wording to commit

What you take away

A working environment and a first CLAUDE.md for your own project.

Setup checklist and CLAUDE.md template

  • Yes: git, Node LTS, an editor and the terminal you use daily are installed
  • Yes: An agent (Claude Code or Codex CLI) is installed, signed in, version checked
  • Yes: The project root has CLAUDE.md or AGENTS.md: stack, how to run, how to verify
  • Yes: CLAUDE.md lists the commands: dev, build, test, typecheck — exactly as in package.json
  • Yes: Prohibitions are written down: what the agent must not touch (migrations, prod configs, secrets)
  • Yes: .gitignore covers .env*, the build folder and local agent settings
  • Yes: The first task is done and committed, the diff reviewed by eye

Harness in this episode

The harness starts here: CLAUDE.md is its first layer. The agent knows how to verify its work before it starts.

The full harness episode

What to have ready

  • Yes: A laptop with admin rights
  • Yes: A Claude account (Pro/Max) or an API key
  • Yes: Any small project in git — an empty one is fine

Official documentation