learn.vdmitriev.com
Tema

Claude Code desde cero

Desde instalar Claude Code hasta construir tu propio agente. Cuatro módulos, cuestionarios, práctica verificada en tu terminal, sin registro.

Módulo 1. Cómo usar Claude Code

Instalación, primera tarea, CLAUDE.md, modos de permisos, contexto y coste. Sin programar.

  1. 01Getting Started with Claude Code Sin empezarClaude 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 minutos
  2. 02Prompt Tips for Claude Code Sin empezarThe 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 minutos
  3. 03Writing Effective CLAUDE.md Files Sin empezarCLAUDE.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 minutos
  4. 04Permission Modes Sin empezarClaude 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 minutos
  5. 05Build a landing page with Claude Code Sin empezarHands-on practice. Describe your product, let Claude Code build a one-page site, review it and ship it.45 minutos
  6. 06Managing the Context Window Sin empezarClaude 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 minutos

Módulo 2. Un agente: bucle, herramientas, control

Cómo funciona por dentro un agente de código, paso a paso en Python.

  1. 01The Agent Loop Sin empezarOne Loop Is All You Need5 minutos
  2. 02Tool Use Sin empezarAdd a Tool, Add Just One Line5 minutos
  3. 03Permission Sin empezarCheck Permissions Before Execution5 minutos
  4. 04Hooks Sin empezarHang on the Loop, Don't Write into It10 minutos
  5. 05TodoWrite Sin empezarAn Agent Without a Plan Drifts Off Course5 minutos
  6. 06Subagent Sin empezarGive a Subtask Its Own Context5 minutos

Módulo 3. Contexto y tareas largas

Skills, compactación del contexto, memoria, sistema de tareas y trabajo en segundo plano.

  1. 01Skill Loading Sin empezarLoad Skills When Needed5 minutos
  2. 02Context Compact Sin empezarMake Room Before the Context Fills Up15 minutos
  3. 03Memory Sin empezarKeep Useful Knowledge Across Sessions10 minutos
  4. 04Task System Sin empezarFrom an Execution Checklist to Coordinated Task State10 minutos
  5. 05Background Tasks Sin empezarSlow Operations Go to the Background10 minutos

Módulo 4. Autonomía y equipos de agentes

Planificador, equipos de agentes, plugins MCP, un arnés integrado, flujos de trabajo y el bucle de objetivos.

  1. 01Cron Scheduler Sin empezarStart Work on a Schedule10 minutos
  2. 02Agent Teams Sin empezarRuntime and Coordination Protocols20 minutos
  3. 03MCP Tools Sin empezarDiscover and Invoke External Tools10 minutos
  4. 04Integrated Harness Sin empezarMany Mechanisms, One Loop15 minutos
  5. 05Workflow Runtime Sin empezarThe Model Decides Each Step; a Script Decides the Orchestration15 minutos
  6. 06Goal Loop Sin empezarThe Model Proposes a Stop; an Independent Evaluator Decides Whether to Continue10 minutos