Quick Reference: Domain 3 — Claude Code

The guide names Rules, Skills, Commands, Agents and Agent Memory. Each one is a file you can open, and nearly every question in this domain is a placement question: given a stated need, which kind of file and which scope.

Collisions. A skill and a command file of the same name both create /name, and the skill takes priority. Across scopes, skill collisions resolve enterprise over personal over project, the exact inverse of settings precedence, and a common trap. Subagent name clashes run the same way as settings precedence instead: managed, then --agents JSON, then .claude/agents/, then ~/.claude/agents/, then plugins. Skills and command files both take arguments through $ARGUMENTS, with $0 and $1 indexing positions zero-based, so $0 is the first argument.

Features related to the five components (September 2026).

• skillOverrides: user-invocable-only hides a skill from Claude but keeps /name working, name-only lists it without its description, off hides it everywhere. The /skills menu saves the choice to .claude/settings.local.json.
• Subagent model: a definition's model, or one Claude passes at spawn time, outranks CLAUDE_CODE_SUBAGENT_MODEL. Add CLAUDE_CODE_SUBAGENT_MODEL_FORCE=1 to put every subagent on that model. This needs v2.1.257 or later. Forks and model: inherit skills still use the main model.
• Subagent permissionMode: ignored while the main session is in acceptEdits, auto or bypassPermissions; the subagent takes the parent's mode. To stop a subagent editing in every mode, leave editing tools out of its tools list.
• Plugins: skills, subagents, hooks and MCP servers packaged as one unit. A repository shares a private marketplace's plugin through extraKnownMarketplaces and enabledPlugins in its .claude/settings.json; the marketplace entry applies only in a folder the contributor has trusted. Enabling the plugin there does not install it: each contributor runs claude plugin install <name>@<marketplace> --scope project once, unless a local directory or file marketplace lists it by relative path. Managed settings carry the same two keys for the whole organisation.
• Output styles: .claude/output-styles/ or ~/.claude/output-styles/, chosen with outputStyle. On the full system prompt, a custom style drops Claude Code's software engineering instructions unless its frontmatter sets keep-coding-instructions: true; the shorter system prompt never carries them.

The sentence that answers most placement questions: CLAUDE.md is what you write for Claude, auto memory is what Claude writes for itself. Both are context. Neither enforces anything.

Four scopes, broadest first:

1. Managed policy: /Library/Application Support/ClaudeCode/CLAUDE.md (macOS), /etc/claude-code/CLAUDE.md (Linux and WSL), C:\Program Files\ClaudeCode\CLAUDE.md (Windows). Deployed by IT, not excludable by individual settings.
2. User: ~/.claude/CLAUDE.md, which applies in every project.
3. Project: ./CLAUDE.md or ./.claude/CLAUDE.md, shared through version control.
4. Local: ./CLAUDE.local.md, personal and gitignored.

Discovered files are concatenated into context, never overridden: a more specific file appears after a broader one rather than replacing it. Claude Code searches upward through the directory tree from the working directory. So a session launched in foo/bar/ reads foo/CLAUDE.md before foo/bar/CLAUDE.md. It reads each CLAUDE.local.md after the CLAUDE.md in the same directory. Files in subdirectories below load on demand when Claude reads there.

• @path/to/import pulls in content from anywhere in a CLAUDE.md. Relative paths resolve against the file containing the import, imports expand at launch, and recursion stops at a maximum depth of four hops. A path in backticks stays literal.
• /init is repository initialisation: Claude analyses the codebase and writes a starting CLAUDE.md of build commands, test instructions and conventions. Where one already exists it suggests improvements rather than overwriting.
• /context proves what actually loaded, under Memory files. /memory opens any of it for editing.
• claudeMdExcludes (September 2026) skips CLAUDE.md and rules files by path or glob, from any settings file; managed policy files cannot be excluded. In .claude/settings.local.json it keeps another team's file out of your sessions only.
• AGENTS.md (v2.1.277 or later) loads by default only when no CLAUDE.md, .claude/CLAUDE.md or CLAUDE.local.md exists in the working directory or above, so a personal CLAUDE.local.md switches it off. Project instructions set to claude-md-and-agents-md in /config loads both; Claude Code ignores that value in project and local settings files.

Settings run the other way, as a genuine precedence chain:

Local, then project, then user is the priority order that questions test for a single value such as model, and candidates make mistakes at the middle step, between local and project: being checked in gives a file no priority. List settings such as permissions.allow merge across files (September 2026), so a local file can add entries but never remove a shared one. To prompt for a command the team allows, add an ask rule: rules are checked deny, then ask, then allow. Claude Code auto-gitignores the local file by adding **/.claude/settings.local.json to your global git excludes file, which excludes it from version control and not from loading.

Settings carry permissions (allow, ask, deny, defaultMode, additionalDirectories), hook configuration, env, model selection, and toggles such as autoMemoryEnabled.

Workspace trust (September 2026). A cloned repository cannot give itself permissions. The permissions.allow rules and additionalDirectories in its .claude/settings.json take effect only once someone accepts the trust dialog for that folder. The same applies to .mcp.json approvals such as enabledMcpjsonServers committed there. Deny and ask rules apply regardless.

Status line. statusLine runs your script with session JSON on stdin. It re-runs the script on events such as a new assistant message. So the status line can stop updating while the main session is idle. refreshInterval adds a timer. Store any cache under the input's session_id, not a process ID, which changes on every run.

The sentence that questions are built from, word for word from the memory page: "Settings rules are enforced by the client regardless of what Claude decides to do. CLAUDE.md instructions shape Claude's behavior but are not a hard enforcement layer." When a question says must never, the right answer is configuration, not prose.

Sessions save continuously as local JSONL transcripts at ~/.claude/projects/<project>/<session-id>.jsonl, so exiting loses nothing.

Sessions created by claude -p or the Agent SDK stay out of both the picker and a plain claude --continue (September 2026). They still resume by ID, and claude -p --continue includes them. A resumed session brings back the conversation and the model it was using. The permission mode has stricter rules (September 2026). It returns only on a terminal resume by --continue, by ID or by a unique name without -p. Even then, it does not return for bypassPermissions or plan (among a few narrower cases). In those cases, the session gets the mode that a new session would start in. The picker and /resume never restore it. claude -p --resume or -p --continue starts in the same mode as a new -p run, except in one narrow plan-mode case. A resumed session does not keep launch flags either: pass --add-dir, --mcp-config, --settings, --plugin-dir or --fallback-model again, or move the setting into a settings file, which is re-read at every launch.

Inside a session: /clear starts an empty context for unrelated work, and the old conversation stays on disk and resumable. /compact [instructions] replaces history with a summary so the same task continues. /context shows what is using the context window. Treating the two as interchangeable ways to free space is a common trap: only one of them lets the same task continue.

claude -p (--print) is the non-interactive mechanism: you give it a prompt, and it returns a result. It accepts piped stdin up to 10MB. It exits 0 on success and non-zero on failure, so scripts can branch. It starts in the Manual permission mode on every plan, so scripted runs set which actions they may take without asking.

Dated current-state note (September 2026): --bare skips auto-discovery of hooks, skills, custom commands, subagents, installed plugins, MCP servers, auto memory and CLAUDE.md. The docs recommend it for scripted and SDK calls and expect it to become the default for -p. A bare run gets what it needs by flag, such as --agents or --mcp-config. A -p run never shows the trust dialog. In a folder nobody has trusted, it ignores the project's permissions.allow rules with a stderr warning, yet without --bare it still runs the hooks in the project's settings files and connects its .mcp.json servers. So the allow rules are ignored, and the hooks run and the servers connect. For a checkout you did not write, --bare or --setting-sources user keeps both out, and the allowlist goes on the command line with --allowedTools.

Invented flags with names close to real ones are a common wrong option here: --headless, --stream and --auto do not exist.

The guide's auto-mode feature. Config values are exact, and values that are almost right are the wrong options:

Two rules outlive every mode: deny rules block even under bypassPermissions, and allow rules have no effect there.

Shift+Tab cycles modes mid-session and permissions.defaultMode sets where sessions start, with two exceptions in .claude/settings.json and .claude/settings.local.json. An auto value there does not take effect, and Claude Code falls back to the built-in default. Nor does bypassPermissions, which starts the session in Manual. Dated current-state note (September 2026): from Claude Code v2.1.283, interactive terminal and VS Code sessions start in auto mode on every plan and provider, while claude -p starts in Manual on every plan. A managed defaultMode only sets the starting point, and people can still switch to auto. disableAutoMode set to "disable" in managed settings makes it unavailable.

The Bash sandbox (September 2026) is a separate layer: the mode decides whether Claude asks, and the sandbox has the operating system limit what a command and its child processes can reach. sandbox.filesystem.allowWrite opens one extra writable path, such as a build cache, which the docs prefer to exempting the tool through excludedCommands. sandbox.allowUnsandboxedCommands: false stops Claude from retrying a blocked command outside the sandbox.

The judgement that questions test: give Claude more freedom to act without asking when the work is more isolated and easier to reverse. acceptEdits for edits you will read in a diff, plan before touching code you do not know, dontAsk with an explicit allowlist for CI, auto for an unattended run whose commands cannot be listed in advance, bypassPermissions only where the whole environment is disposable.