Zero to expert Claude Code. Install, CLAUDE.md, the .claude/ ecosystem, permissions, slash commands, hooks, subagents, MCP, git workflows, code review, web grounding, background tasks, and team-scale deployment.
Install Claude Code, log in (Pro/Max/API/Enterprise), pick the right plan for your usage, and run your first session.
Understand what Claude Code actually is — an agentic CLI that reads your codebase, runs commands, and writes files. The plan → act → verify loop.
Write prompts that get results. Plan mode, task decomposition, specificity, and the art of describing intent for an agent.
The single highest-leverage file in your repo. Tech stack, conventions, do/don't rules, and how to make Claude actually follow them.
Settings, agents, commands, skills, hooks, plugins, output styles, statusline. Map every file in .claude/ and what it controls.
Five permission modes, Bash allowlists/denylists, --dangerously-skip-permissions, and the four safety failure modes.
Built-in slash commands, custom commands in .claude/commands/, skills as reusable capabilities, and when each pattern wins.
Wire shell, HTTP, prompt, and agent hooks to PreToolUse, PostToolUse, SessionStart/End, and the 2026 conditional `if` field.
Define reusable specialized agents in .claude/agents/. Sub-agent vs fork, common recipes (code-reviewer, docs-writer, test-generator).
Register MCP servers (GitHub, Slack, Postgres, Linear, Figma...) — auth patterns, /mcp command, alwaysLoad option, concurrent connections.
Build features, APIs, components, and tests from natural language. The vertical slice workflow.
Paste errors, read stack traces, fix linting, refactor safely. Claude Code as your debugging partner.
Conventional commits, branches, PRs, and worktrees for parallel agentic work. Settings.json patterns to prevent the four classic git incidents.
Fast local /review and multi-agent cloud /ultrareview. REVIEW.md customization, GitHub Actions / GitLab CI integration for automated PR feedback.
WebFetch known URLs, WebSearch for discovery, Context7 for library docs. When to ground vs trust training, and the four web-grounding pitfalls.
run_in_background, Monitor for streaming events, ScheduleWakeup for one-shot deferred work, CronCreate for recurring autonomous agents.
Onboarding 50 engineers, cost management, billing tiers, enterprise managed settings, audit logs, and scaling Claude Code across an org.
425 questions across 17 modules — check how well you understood this track.