Claude / Operating Manual

The Claude Code operating manual.

The terse-lookup layer of the Claude Manual. One page per capability, written for people who already know what they are after and want the shape of it fast. If you are new to Claude, read Start Here first.

Topics

Deep guides organized by what you are trying to do. The best place to start when you have a goal, not a keyword.

  1. Primitives the principles assumeThe generic Claude primitives the KMS principles build on, as one digest: subagents, skills, CLAUDE.md, memory, hooks, MCP, long context, prompt caching, fast mode, tool selection, context window management.
  2. Loops & AutonomyMake Claude Code run itself. In-session orchestration, self-paced multi-turn loops, scheduled cloud routines, and spec-driven autonomous runs. When to reach for which, and the cadence math nobody tells you.
  3. Memory & ContextClaude Code only remembers what you write down, and everything you write down loads every turn. How to keep the right things present and the rest out, across sessions and within one.
  4. Orchestration & Agent TeamsOne agent in one context window is serial and fills up. Orchestration buys you parallelism and fresh context. Subagents, agent teams, and workflows are three different tools, and the difference is how the agents coordinate.
  5. Models & CostModel choice is the biggest cost lever you have, roughly a 30x spread for the same work. Default to the cheapest model that clears the task, override per project, and reach up only for the hard parts.

Reference

Quick reference for each individual capability, one page each.

  1. SubagentsSpawn specialized agents for parallel or isolated work in their own context.
  2. Agent teamsFull peer agent sessions that message each other and share a task list, for work that needs live coordination rather than fire-and-forget delegation.
  3. SkillsProgressive-disclosure knowledge units, loaded by name at startup, with full instructions pulled only when Claude matches them to the task.
  4. Background TasksCoordinate long-running work across sessions, subagents, and context windows with a shared task list stored on disk.
  5. Scheduled tasks & /loopRun a prompt on a repeat or on a clock. /loop and the Cron tools schedule inside a session; /schedule runs detached cloud routines.
  6. MCP ServersConnect Claude Code to external tools and data sources via the Model Context Protocol.
  7. HooksShell commands that fire at lifecycle events. Use them to validate, block, log, or format without touching Claude's logic.
  8. Slash CommandsControl sessions and automate workflows with built-in and custom `/` commands.
  9. PluginsSingle-command installable bundles of slash commands, agents, MCP servers, and hooks.
  10. Memory & CLAUDE.mdInstruction files that load every session, giving Claude persistent project and personal context without re-explaining yourself each time.
  11. Context managementSteer what stays in the window and what persists across sessions. /compact directs compaction, /clear resets, /memory edits the memory files.
  12. WorkflowsHave Claude orchestrate tens to hundreds of parallel subagents in the background, with verification.
  13. Remote & WebRun and steer Claude Code beyond one terminal: remote sessions, cross-device Remote Control, and the desktop GUI.

Setup

Install, config, permissions, model selection.

  1. Installation & SetupGet Claude Code installed, authenticated, and ready to use in one sitting.
  2. Permissions & SettingsConfigure what Claude can run without asking via settings.json allowlists at project or user scope.
  3. Model SelectionWhich Claude model to run, when to flip to fast mode, and how fallbacks work.
  4. IDE & VS CodeRun Claude Code inside VS Code or JetBrains instead of a standalone terminal.

Releases

Running changelog of models and Claude Code drops.

  1. ChangelogRunning log of notable Claude Code drops and model releases.