⌨Keyboard Shortcuts
Core
⌃CCtrl+CCancel input / generation
⌃DCtrl+DExit session
⌃LCtrl+LClear screen
⌃OCtrl+OToggle verbose output
⌃RCtrl+RReverse history search
⌃GCtrl+GOpen prompt in editor
⌃BCtrl+BBackground task
⌃TCtrl+TToggle task list
⌘VCtrl+VPaste image
⌃F ×2Ctrl+F ×2Finish background agents
Esc EscUndo / revert
Mode Switching
⇧TabCycle permission modes
⌥PAlt+PSwitch model
⌥TAlt+TToggle thinking
Input
⌥Enter\+EnterNew line (quick)
⌃JCtrl+JNew line (control seq)
Prefixes
/Slash command
!Direct bash command
@File mention + autocomplete
⬡MCP Servers
Transports
--transport httpRemote HTTP (recommended)
--transport stdioLocal process
--transport sseRemote SSE
Scopes
Local.claude.json (project)
Project.mcp.json (shared / VCS)
User~/.claude.json (global)
Management
/mcpInteractive UI
claude mcp listList all servers
claude mcp serveRun CC as MCP server
⚡Slash Commands
Session
/clearClear conversation
/compact [focus]Compress context
/resumeResume / switch session
/rename [name]Name current session
/branch [name]Fork conversation (alias /fork)
/costToken usage stats
/contextContext usage visualizer
/diffInteractive diff viewer
/copyCopy last response
/exportExport conversation
/rewindCheckpoint / rewind to previous point
/statusShow version, model, account status
/statsDaily usage, streaks, preferences
/tasksList background tasks
Settings
/configOpen settings
/model [model]Switch model (←→ effort)
/fast [on|off]Toggle fast mode
/vimToggle vim mode
/themeChange color theme
/permissionsView / update permissions
/effort [level]Set effort (low / med / high)
/color [color]Prompt line color
/keybindingsCustomize keybindings
/terminal-setupConfigure terminal keys
/ideManage IDE integrations
/privacy-settingsView / update privacy settings
/remote-envConfigure web session environment
/desktopContinue in Desktop app
Tools
/initCreate CLAUDE.md
/memoryEdit CLAUDE.md
/mcpManage MCP servers
/hooksManage hooks
/skillsList available skills
/agentsManage agents
/add-dir <path>Add working directory
/reload-pluginsHot-reload plugins
/pluginManage Claude Code plugins
/install-github-appSet up GitHub Actions app
/install-slack-appInstall Slack app
/web-setupConnect GitHub to web
Special
/btw <question>Side question (no context cost)
/plan [desc]Plan mode (+ auto-start)
/ultraplan <prompt>Draft plan in browser, execute remotely
/powerupInteractive lessons with animated demos
/teleportPull web session to terminal
/autofix-pr [prompt]Auto-fix PR on CI failures
/loop [interval]Schedule recurring task
/voiceVoice input (20 languages)
/doctorDiagnose installation
/pr-comments [PR]GitHub PR comments
/remote-controlBridge to claude.ai/code
/usagePlan limits & status
/scheduleCloud scheduled tasks
/security-reviewSecurity analysis
/insightsAnalyze sessions & patterns
/sandboxToggle sandbox mode
/extra-usageConfigure extra usage quota
/helpHelp + commands
/feedbackSend feedback (alias: /bug)
◎Memory & Files
CLAUDE.md Locations
./CLAUDE.mdProject (team-shared)
~/.claude/CLAUDE.mdPersonal (all projects)
/etc/claude-code/Managed (org-wide)
Rules & Imports
.claude/rules/*.mdProject rules
~/.claude/rules/*.mdUser rules
paths: frontmatterPath-specific rules
@path/to/fileImport in CLAUDE.md
Auto Memory
~/.claude/projects/<proj>/memory/MEMORY.md + topic files, auto-loaded
◇Workflows & Tips
Thinking & Effort
⌥TAlt+TToggle thinking
"ultrathink"Max effort for one turn
⌃OCtrl+OView thinking (verbose)
/effort○ low · ◐ med · ● high
Plan Mode
⇧TabNormal → Auto-accept → Plan
--permission-mode planLaunch in plan mode
Git Worktrees
--worktree nameIsolated branch for feature
isolation: worktreeAgent in own worktree
sparsePathsCheckout only needed folders
/batchAuto-create worktrees
Context Management
/contextUsage + optimization tips
/compact [focus]Compress with focus
Auto-compact~95% capacity
1M contextOpus 4.6 (Max / Team / Ent)
CLAUDE.mdSurvives compaction
Session Tips
claude -cContinue last conversation
claude -r "name"Resume by name
/btw questionSide question, no context cost
SDK / Headless
claude -p "prompt"Non-interactive mode
--output-format jsonStructured output
--max-budget-usd 5Cost limit
cat file | claude -pPipe input
Voice Mode
/voiceEnable push-to-talk
Space (hold)Record; release = send
20 languagesEN, ES, FR, DE, CZ, PL…
⚙Config & Environment
Config Files
~/.claude/settings.jsonUser settings
.claude/settings.jsonProject (shared)
.claude/settings.local.jsonLocal only
~/.claude.jsonOAuth, MCP, state
.mcp.jsonProject MCP servers
Key Settings
modelOverridesMap model selection → ID
autoMemoryDirectoryCustom memory directory
worktree.sparsePathsSparse checkout folders
Environment Variables
ANTHROPIC_API_KEY
ANTHROPIC_MODEL
CLAUDE_CODE_EFFORT_LEVELlow / med / high
MAX_THINKING_TOKENS0 = off
CLAUDE_CODE_MAX_OUTPUT_TOKENSdefault 32K
CLAUDE_CODE_DISABLE_CRONDisable scheduled tasks
✦Skills & Agents
Built-in Skills
/simplifyCode review (3 parallel agents)
/batchBulk changes (5–30 worktrees)
/debug [desc]Debug from log
/loop [interval]Recurring scheduled task
/claude-apiLoad API + SDK reference
Custom Skill Locations
.claude/skills/<name>/Project skills
~/.claude/skills/<name>/Personal skills
Skill Frontmatter
descriptionAuto-invoke trigger
allowed-toolsSkip permission prompts
modelOverride model for skill
effortOverride effort level
context: forkRun in subagent
$ARGUMENTSUser input placeholder
${CLAUDE_SKILL_DIR}Skill directory
!`cmd`Dynamic context injection
Built-in Agents
ExploreFast read-only (Haiku)
PlanResearch for plan mode
GeneralAll tools, complex tasks
BashTerminal, separate context
Agent Frontmatter
permissionModedefault/acceptEdits/plan/dontAsk/bypass
isolation: worktreeRun in git worktree
memory: user|projectPersistent memory
background: trueBackground task
maxTurnsLimit agent turns
SendMessageResume stopped agents
▸CLI & Flags
Core Commands
claudeInteractive mode
claude "prompt"With prompt
claude -p "prompt"Headless
claude -cContinue last
claude -r "name"Resume by name
claude updateUpdate
Key Flags
--modelSet model
-wGit worktree
-n / --nameSession name
--add-dirAdd directory
--agentUse an agent
--allowedToolsPre-approve tools
--output-formatjson / stream
--json-schemaStructured output
--max-turnsLimit turns
--max-budget-usdCost limit
--consoleAuth via Anthropic Console
--verboseVerbose output
--bareMinimal headless (no hooks/LSP)
--channelsPermission relay / MCP push
--remoteWeb session
--effortlow / med / high / max
--permission-modeplan / default / …
--dangerously-skip-permissionsSkip all prompts ⚠