agents101 · Best Practices

About

The mission, content sources, and module overview of agents101.

Mission

agents101 provides systematic best practices for developers using AI agent tools — Claude Code, Codex, OpenClaw, and Hermes. We integrate open-source community knowledge, expert experience, and in-depth reports into one place, helping developers go from beginner to expert.

Content Sources

This site is built on the following open-source projects and community resources:

  • claude-code-best-practice — Primary content source: best practices, in-depth reports, expert tips
  • Anthropic Official Documentation — Official Claude Code usage guide
  • Community videos and podcasts — Public talks by Boris Cherny, Karpathy, Matt Pocock, and others

Modules

The Claude Code guide is organized into 8 modules:

  • Getting Started — Installation, CLI args, Power-ups, quick start path
  • Core Concepts — Architecture, Command-Agent-Skill pipeline, Subagents
  • Skills & Hooks — 9 skill types, Hooks lifecycle, advanced tool usage
  • Config & Memory — Settings, permissions, CLAUDE.md, memory system
  • Workflows — RPI, CRISPY, cross-model collaboration, agent teams
  • Deep Reports — LLM degradation, harness importance, learning paths
  • Expert Insights — Boris Cherny's rules, Thariq's tips, categorized collection
  • Cheat Sheet — Quick reference for commands, flags, env vars, templates

Languages

Content is available in English, 简体中文, 日本語, 한국어, and Español.