What Claude Code Is
The agentic CLI/IDE tool explained — the session loop, install, and where everything lives.
CLAUDE.md & Memory Files
The highest-leverage Claude Code customization: persistent project memory.
Plan Mode
Read-only investigate-and-propose before any edits run — the safest way to tackle big changes.
IDE Integrations
VS Code and JetBrains extensions, inline diffs, and shared settings.
settings.json: The Config System
File tiers and precedence, key sections, and project vs personal config.
Permissions & Permission Modes
allow/ask/deny rules and modes — cut prompts without losing safety.
Custom Slash Commands
Turn a repeatable workflow into one word.
Output Styles
Change Claude Code's tone, verbosity, and teaching behaviour.
Statusline Customization
Surface model, cwd, context budget and git state with a custom script.
Context Management
/compact vs /clear, auto-compaction, and staying effective as sessions grow.
Hooks: Deterministic Automation
Run commands at lifecycle events to enforce formatting, gates, and safety.
Skills: On-Demand Expertise
SKILL.md, progressive disclosure, and Skill vs command vs subagent vs MCP.
Subagents & Parallel Agents
Delegate to isolated-context agents with scoped tools — and when not to.
MCP Servers in Claude Code
Connect databases, GitHub, browsers and more via the Model Context Protocol.
Background Tasks, /loop & Scheduling
Long-running commands, recurring loops, and scheduled cloud agents.
Plugins & Marketplaces
Bundle skills, agents, commands, hooks and MCP into one installable unit.
Git Worktrees & Parallel Workstreams
Run multiple isolated Claude sessions on one repo without them colliding.
Headless Mode & the Agent SDK
Run Claude Code in CI, and build production agents on the same harness.