Domain 3: Claude Code

3.1% of exam

ts-ccdvf-3.1

Identify Claude Code's core component types

Key Points

Five component types: Rules (standing conventions, often in CLAUDE.md), Skills (SKILL.md, loaded when relevant), Commands (built-in + custom slash commands in .claude/commands/), Agents (delegated subagents), Agent Memory (persisted context, notably CLAUDE.md).
Skill vs. Command distinction: Skills load implicitly on relevance; Commands are explicitly invoked by name.
CLAUDE.md is the primary instance of Agent Memory, not a separate mechanism.
Built-in slash commands include /clear (reset context), /init (bootstrap CLAUDE.md), /help.

Decision Rules

When: A capability is invoked explicitly by name (/foo)

→It's a Command, whether built-in or custom.

When: A capability loads automatically because Claude judges it relevant to the task

→It's a Skill, defined by a SKILL.md.

When: A subtask is delegated to a focused-role, focused-tool-set helper

→It's an Agent (subagent), not a Command.

✗ Anti-Patterns to Reject

Confusing Skills with Commands because both are Markdown-authored, reusable instructions -- the distinguishing factor is invocation.
Treating Agent Memory and CLAUDE.md as two separate mechanisms.

ts-ccdvf-3.2

Apply the CLAUDE.md memory hierarchy

Key Points

CLAUDE.md is auto-loaded project memory, not executable configuration.
Four hierarchy levels, broadest to most specific: enterprise/system, user (~/.claude/CLAUDE.md), project (./CLAUDE.md), subdirectory.
More specific files layer on top of broader ones -- they refine, they don't replace.
Project-level CLAUDE.md is checked into version control so the whole team shares the same guidance.
/init bootstraps a starter CLAUDE.md from an existing codebase.

Decision Rules

When: A team wants the same conventions applied for everyone on a repo

→Use a project-level CLAUDE.md committed to version control.

When: Bootstrapping CLAUDE.md for an existing repository

→Run /init.

When: Permission rules, hooks, or tool allow/deny lists need a home

→Put them in settings.json, never in CLAUDE.md.

✗ Anti-Patterns to Reject

Assuming the user-level ~/.claude/CLAUDE.md is what a team shares -- it is personal, not committed to the repo.
Storing permission rules or hooks in CLAUDE.md.

ts-ccdvf-3.3

Configure behavior and permissions via settings.json

Key Points

settings.json (user: ~/.claude/settings.json, project: .claude/settings.json) configures tool allow/deny lists, hooks, environment variables, model selection, and MCP servers.
It is the deterministic control surface -- an allow/deny rule or hook blocks a dangerous command regardless of what the model proposes.
CLAUDE.md is context/memory (probabilistic influence on the model); settings.json is executable configuration (deterministic enforcement).

Decision Rules

When: A rule must block a destructive shell command regardless of model behavior

→Put it in settings.json as a permission rule or hook, not in CLAUDE.md.

When: Configuring tool allow/deny lists, hooks, environment variables, model choice, or MCP servers

→settings.json is the surface, at the correct scope (user vs. project).

✗ Anti-Patterns to Reject

Putting permission rules, hooks, or tool allow/deny lists in CLAUDE.md.
Assuming a well-written CLAUDE.md instruction is an adequate substitute for a settings.json permission rule.

ts-ccdvf-3.4

Distinguish Claude Code's session and operating-mode features

Key Points

Session management: conversations persist as resumable sessions; /clear resets context to stay focused.
Headless mode runs non-interactively (e.g., claude -p) for scripting/CI, with no interactive TUI.
Streaming mode emits incremental output, including structured stream-JSON, for programmatic consumption.
Auto-mode is reduced-friction autonomous operation with fewer confirmations -- pair it with settings.json permission rules and hooks.
Headless mode (interactivity) and auto-mode (autonomy) are independent axes, frequently confused.
Best-practice loop: gather context -> plan -> act -> verify, tightened with custom commands and hooks rather than long, brittle prompts.

Decision Rules

When: Running Claude Code non-interactively in a CI pipeline

→Use headless mode (e.g., claude -p).

When: Output must be consumed incrementally or programmatically

→Use streaming mode.

When: Using auto-mode

→Pair it with settings.json permission rules and hooks -- never use it unguarded.

✗ Anti-Patterns to Reject

Confusing headless mode (interactivity) with auto-mode (autonomy) -- they are independent axes.
Using auto-mode without corresponding permission rules or hooks.