learn.vdmitriev.com
Theme

Claude Code from zero

From installing Claude Code to building your own agent. Four modules, quizzes, practice checked in your terminal, no sign-up.

Module 1. How to use Claude Code

Install, first task, CLAUDE.md, permission modes, context and cost. No programming required.

  1. 01Getting Started with Claude Code Not startedClaude Code is an agentic coding assistant that operates directly in your terminal. It can read and edit files, run commands, search your codebase, and interact with Git and GitHub -- all through natural language. This guide walks you through installation, authentication, and your first productive session.5 minutes
  2. 02Prompt Tips for Claude Code Not startedThe quality of Claude Code's output depends heavily on how you communicate your intent. This guide covers practical techniques for writing clear, effective prompts that get the results you want on the first try.5 minutes
  3. 03Writing Effective CLAUDE.md Files Not startedCLAUDE.md is a special Markdown file that gives Claude Code persistent context about your project. Think of it as a briefing document -- it tells Claude your conventions, tech stack, testing commands, and preferences so you don't have to repeat them every session.5 minutes
  4. 04Permission Modes Not startedClaude Code operates within a permission system that controls which tools and actions it can take without asking for approval. Understanding these modes lets you balance safety with speed depending on your task and comfort level.15 minutes
  5. 05Build a landing page with Claude Code Not startedHands-on practice. Describe your product, let Claude Code build a one-page site, review it and ship it.45 minutes
  6. 06Managing the Context Window Not startedClaude Code operates within a context window whose size depends on your model and plan. Opus 4.8 supports a 1M token context window by default on Max, Team, and Enterprise plans. Sonnet and Haiku have smaller windows. How you manage that window directly affects response quality, speed, and cost. This guide covers strategies for keeping context focused and knowing when to start fresh.20 minutes

Module 2. One agent: loop, tools, control

How a coding agent works inside, built step by step in Python.

  1. 01The Agent Loop Not startedOne Loop Is All You Need5 minutes
  2. 02Tool Use Not startedAdd a Tool, Add Just One Line5 minutes
  3. 03Permission Not startedCheck Permissions Before Execution5 minutes
  4. 04Hooks Not startedHang on the Loop, Don't Write into It10 minutes
  5. 05TodoWrite Not startedAn Agent Without a Plan Drifts Off Course5 minutes
  6. 06Subagent Not startedGive a Subtask Its Own Context5 minutes

Module 3. Context and long tasks

Skills, context compaction, memory, task system and background work.

  1. 01Skill Loading Not startedLoad Skills When Needed5 minutes
  2. 02Context Compact Not startedMake Room Before the Context Fills Up15 minutes
  3. 03Memory Not startedKeep Useful Knowledge Across Sessions10 minutes
  4. 04Task System Not startedFrom an Execution Checklist to Coordinated Task State10 minutes
  5. 05Background Tasks Not startedSlow Operations Go to the Background10 minutes

Module 4. Autonomy and agent teams

Scheduler, agent teams, MCP plugins, an integrated harness, workflows and the goal loop.

  1. 01Cron Scheduler Not startedStart Work on a Schedule10 minutes
  2. 02Agent Teams Not startedRuntime and Coordination Protocols20 minutes
  3. 03MCP Tools Not startedDiscover and Invoke External Tools10 minutes
  4. 04Integrated Harness Not startedMany Mechanisms, One Loop15 minutes
  5. 05Workflow Runtime Not startedThe Model Decides Each Step; a Script Decides the Orchestration15 minutes
  6. 06Goal Loop Not startedThe Model Proposes a Stop; an Independent Evaluator Decides Whether to Continue10 minutes