
Already know some of this? A 2-minute test skips the lessons you know.
Take the placement testA free 66-lesson course built from freeCodeCamp.org's 12h 20m of YouTube video. 7 days at 90 minutes a day, with study notes and a quiz in every lesson. A 2-minute placement test skips what you already know.
Taught by freeCodeCamp.org on YouTube. Videos play in YouTube's own player, so the creator keeps every view.
Understand the Cloud Code Essentials certification structure, exam format, target audience, and learning roadmap including prerequisite and follow-up courses.
Define agentic coding tools and code harnesses, compare major tools (Claude Code, OpenAI Codex, Kiro, OpenCode), and understand how they integrate with development environments.
Explain Claude Code as a CLI-based code harness, its use cases, and the three-phase agentic loop (gather context, take action, verify results) with interrupt capability.
Describe the five built-in tool categories, how the agentic loop uses tools, and how to choose between Opus, Sonnet, and Haiku models including the Opus-plan hybrid mode.
Manage Claude subscription usage, understand 5-hour rolling windows, session vs subscription limits, API key vs subscription authentication, and handle 401 token expiry.
Install Claude Code via official CLI scripts on Linux/macOS/WSL2, understand Amazon Linux/EC2 limitations, and authenticate in containerized environments.
Control sessions with /resume, /fork, rename, and rewind commands; distinguish web vs local sessions; understand 30-day session retention.
Explain context window (200k-1M tokens), 5-hour rolling usage limits, model-specific limits, and use /cost, /stats, and ccusage tool to monitor spend.
Customize the CLI status line with bash/Python scripts, queue side questions via 'by the way', and understand voice command limitations on WSL2.
Export sessions to clipboard or file via /export, navigate ~/.claude/projects/ for JSONL session files, and analyze raw event data for usage debugging.
Layer settings across managed (org), user (~/.claude/settings.json), project (.claude/settings.local.json), and local (.claude/settings.local.json) scopes with precedence rules.
Configure tool permissions (bash, read, edit, web fetch, MCP) with deny/ask/allow precedence, wildcard patterns for bash commands, and path syntax (//, ~/, ./).
Enable sandbox mode (bubblewrap, socat, seccomp) to isolate bash tools, test deny rules (e.g., npx), and observe sandbox escape via dangerously-skip-permissions.
Run headless tasks with -p flag for CI/CD, use allow-tools/deny-tools per task, and build custom agents with the Agent SDK (Python/TypeScript) including hooks and MCP.
Trigger PR reviews via @claude review, run simplify on repos, and schedule recurring reminders with natural language or cron via the loop command.
Distinguish CLAUDE.md (docs), rules (enforced), imports (context), and auto-memory (learned preferences); troubleshoot with /memory and /context commands.
Add MCP servers via single-line config in settings.json with env var expansion, discover plugins (skills, agents, hooks, MCP) in marketplace, and understand LSP.
Enable agent teams via environment variable, coordinate lead/teammates with shared task list and mailbox, and run parallel sub-agents for multi-language porting.
Compare desktop app (co-work, visual diff), web (GitHub sandbox), VS Code extension (plan preview), JetBrains plugin (terminal-only), and Chrome extension (remote control).
Use mobile app for remote session access, authenticate via cloud auth login, and control local Claude Code sessions from phone with remote control feature.