[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-83518":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":22,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":45,"readmeContent":46,"aiSummary":10,"trendingCount":16,"starSnapshotCount":16,"syncStatus":15,"lastSyncTime":47,"discoverSource":48},83518,"claude-code-guide","zebbern\u002Fclaude-code-guide","zebbern","Claude Code Guide - Setup, Commands, workflows, agents, skills & tips-n-tricks go from beginner to power user!","",null,"Python",4268,432,34,2,0,31,32,93,99.11,"MIT License",false,"main",true,[26,27,28,29,30,31,32,33,34,5,35,36,37,38,39,40,41,42,43,44],"ai","ai-agent","ai-agent-tools","anthropic-claude","claude","claude-ai","claude-api","claude-code","claude-code-communication","claude-code-skills","claude-commands","claude-desktop","claude-mcp","claude-sonnet","code","mcp","mcp-agents","mcp-tools","vscode-extension","2026-06-12 04:01:41","\u003Cdiv align=\"center\">\n\n\u003Ch2 id=\"claude-code-community-guide\">Claude Code Guide\u003C\u002Fh2>\n\n_For reference and contributions, visit the [official Claude Code documentation](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Foverview)_\n\n![Claude Code](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002F@anthropic-ai\u002Fclaude-code?label=Claude%20Code&logo=anthropic)\n[![Status](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FStatus-Active-brightgreen)](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-code)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Anthropic-orange)](https:\u002F\u002Fclaude.ai\u002Fcode)\n\n\u003C\u002Fdiv>\n\n\u003Cdiv align=\"center\">\n\n\u003Ckbd>\n\n| Section                               | Status | Other Resources                                                                                         |\n| ------------------------------------- | ------ | ------------------------------------------------------------------------------------------------------- |\n| Getting Started                       | ✅   | **Claude-Code** [Docs](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Foverview)                                    |\n| Configuration & Environment Variables | ✅   | **Claude-Code via** [**Discord**](https:\u002F\u002Fgithub.com\u002Fzebbern\u002Fclaude-code-discord)                       |\n| Commands & Usage                      | ✅   | Security Agents [SKILL.md](https:\u002F\u002Fgithub.com\u002Fzebbern\u002Fclaude-code-guide\u002Ftree\u002Fmain\u002Fskills)               |\n| Interface & Input                     | ✅   | Let Agent Create [SKILL.md](https:\u002F\u002Fgithub.com\u002Fzebbern\u002Fagent-skills-authoring)                          |\n| Advanced Features                     | ✅   | 954+ Agent [Skills](https:\u002F\u002Fgithub.com\u002Fzebbern\u002Fantigravity-awesome-skills)                              |\n| Automation & Integration              | ✅   | No cost ai [resources](https:\u002F\u002Fgithub.com\u002Fzebbern\u002Fno-cost-ai)                                           |\n| Help & Troubleshooting                | ✅   | 250+ Mermaid [templates](https:\u002F\u002Fgithub.com\u002Fzebbern\u002Fmermaid-templates)                                  |\n| Third-Party Integrations              | ✅   | Discord Communication [MCP](https:\u002F\u002Fgithub.com\u002Fzebbern\u002Fdiscord-mcp-agent)                               |\n\n\n\u003C\u002Fkbd>\n\n\u003C\u002Fdiv>\n\n---\n\n\u003Ch3 id=\"content\">Contents\u003C\u002Fh3>\n\n**Fast paths:** [Install](#quick-start) · [Commands](#claude-commands) · [Config](#configuration--environment) · [MCP](#mcp-integration) · [Agents](#sub-agents) · [Troubleshoot](#help--troubleshooting)\n\n| Area | Start here | Also useful |\n| --- | --- | --- |\n| [Getting Started](#getting-started) | [Quick Start](#quick-start) | [Initial Setup](#initial-setup), [System Requirements](#system-requirements) |\n| [Configuration](#configuration--environment) | [Environment Variables](#environment-variables) | [Configuration Files](#configuration-files) |\n| [Commands](#commands--usage) | [Slash Commands](#claude-commands) | [CLI Quick Reference](#cheat-sheet) |\n| [Interface](#interface--input) | [Keyboard Shortcuts](#keyboard-shortcuts) | [Vim Mode](#vim-mode) |\n| [Advanced Features](#advanced-features) | [Plan Mode](#plan-mode), [Auto Mode](#auto-mode), [MCP](#mcp-integration) | [Sub Agents](#sub-agents), [Skills](#skills), [Hooks](#hooks-system) |\n| [Security](#security--permissions) | [Security & Permissions](#security--permissions) | [Dangerous Mode](#dangerous-mode), [Best Practices](#security-best-practices-main) |\n| [Automation](#automation--integration) | [Automation & Scripting](#automation--scripting-with-claude-code) | [PR Review](#auto-pr-review-inline-comments), [Issue Triage](#issue-triage-suggest-labels--severity) |\n| [Help](#help--troubleshooting) | [Troubleshooting](#help--troubleshooting) | [Best Practices](#best-practices), [Monitoring](#monitoring--alerting) |\n| [Third-Party Integrations](#third-party-integrations) | [DeepSeek Integration](#deepseek-integration) | [Provider Setup Examples](#provider-setup-examples) |\n\n\u003Cdetails>\n\u003Csummary>Full content map\u003C\u002Fsummary>\n\n- **[Getting Started](#getting-started)**\n  - [Quick Start](#quick-start)\n  - [System Requirements](#system-requirements)\n  - [Initial Setup](#initial-setup)\n\n- **[Configuration & Environment](#configuration--environment)**\n  - [Environment Variables](#environment-variables)\n  - [Configuration Files](#configuration-files)\n\n- **[Commands & Usage](#commands--usage)**\n  - [Slash Command Reference](#claude-commands)\n  - [CLI Quick Reference](#cheat-sheet)\n\n- **[Interface & Input](#interface--input)**\n  - [Keyboard Shortcuts](#keyboard-shortcuts)\n  - [Vim Mode](#vim-mode)\n\n- **[Advanced Features](#advanced-features)**\n  - [Thinking Mode](#thinking-keywords)\n  - [Effort Levels](#effort-levels)\n  - [Fast Mode](#fast-mode)\n  - [Auto Mode](#auto-mode)\n  - [Plan Mode](#plan-mode)\n  - [Background Tasks](#background-tasks)\n  - [Workflows & Scheduling](#workflows--scheduling)\n  - [Remote Sessions](#remote-sessions)\n  - [Claude in Chrome](#claude-in-chrome)\n  - [Sandbox Mode](#sandbox-mode)\n  - [LSP Tool](#lsp-tool)\n  - [Sub Agents](#sub-agents)\n  - [Agent Teams](#agent-teams)\n  - [Skills](#skills)\n  - [Plugin System](#plugin-system)\n  - [Worktree Isolation](#worktree-isolation)\n  - [Native Installer](#native-installer)\n  - [Authentication CLI](#claude-auth)\n  - [Agent Management CLI](#claude-agents-cli)\n  - [Remote Control](#remote-control)\n  - [Managed Settings](#managed-settings)\n  - [Model Updates](#model-updates)\n  - [Theming & Customization](#theming--customization)\n  - [Code Review](#code-review)\n  - [Insights](#insights)\n  - [MCP Integration](#mcp-integration)\n  - [Hooks System](#hooks-system)\n\n- **[Security & Permissions](#security--permissions)**\n  - [Dangerous Mode](#dangerous-mode)\n  - [Security Best Practices](#security-best-practices-main)\n\n- **[Automation & Integration](#automation--integration)**\n  - [Automation & Scripting](#automation--scripting-with-claude-code)\n  - [Auto PR Review](#auto-pr-review-inline-comments)\n  - [Issue Triage](#issue-triage-suggest-labels--severity)\n\n- **[Help & Troubleshooting](#help--troubleshooting)**\n  - [Installation Issues](#installation--nodejs-issues)\n  - [MCP Issues](#mcp-model-context-protocol-issues)\n  - [Best Practices](#best-practices)\n  - [Monitoring & Alerting](#monitoring--alerting)\n\n- **[Third-Party Integrations](#third-party-integrations)**\n  - [Provider Setup Examples](#provider-setup-examples)\n  - [DeepSeek Integration](#deepseek-integration)\n\n\u003C\u002Fdetails>\n\n---\n\n\u003Ch1 id=\"getting-started\">Getting Started\u003C\u002Fh1>\n\n**Enable sound alerts when claude completes the task:**\n\n> \u003Ckbd>claude config set --global preferredNotifChannel terminal_bell\u003C\u002Fkbd>\n\n\u003Ch2 id=\"quick-start\">Quick Start\u003C\u002Fh2>\n\n> [!TIP]\n> **Send \u003Cmark>claude\u003C\u002Fmark> or \u003Cmark>npx claude\u003C\u002Fmark> in terminal to start the interface**\n>\n> **Go to [Help & Troubleshooting](#help--troubleshooting) to fix issues...**\n\n```C\n# Native Installer (preferred — no Node.js required) ⭐️\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# Windows\n\u002F* Via CMD (native)      *\u002F curl -fsSL https:\u002F\u002Fclaude.ai\u002Finstall.cmd -o install.cmd && install.cmd && del install.cmd\n\u002F* Via Powershell        *\u002F irm https:\u002F\u002Fclaude.ai\u002Finstall.ps1 | iex\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# macOS \u002F Linux \u002F WSL\n\u002F* Via Terminal          *\u002F curl -fsSL https:\u002F\u002Fclaude.ai\u002Finstall.sh | bash\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# Arch\n\u002F* Via Terminal          *\u002F yay -S claude-code*\u002F\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n\n# Alternative (npm) — useful when your environment already standardizes on Node.js\n# Requires Node.js 18+\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# Windows\n\u002F* Via CMD (npm)         *\u002F npm install -g @anthropic-ai\u002Fclaude-code\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# macOS\n\u002F* Via Terminal          *\u002F brew install node && npm install -g @anthropic-ai\u002Fclaude-code\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# Linux\n\u002F* Via Terminal          *\u002F sudo apt update && sudo apt install -y nodejs npm\n\u002F* Via Terminal          *\u002F npm install -g @anthropic-ai\u002Fclaude-code\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# WSL\u002FGIT\n\u002F* Via Terminal          *\u002F npm install -g @anthropic-ai\u002Fclaude-code\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# Docker\n\u002F* Windows (CMD)         *\u002F docker run -it --rm -v \"%cd%:\u002Fworkspace\" -e ANTHROPIC_API_KEY=\"sk-your-key\" node:20-slim bash -lc \"npm i -g @anthropic-ai\u002Fclaude-code && cd \u002Fworkspace && claude\"\n\u002F* macOS\u002FLinux (bash\u002Fzsh)*\u002F docker run -it --rm -v \"$PWD:\u002Fworkspace\" -e ANTHROPIC_API_KEY=\"sk-your-key\" node:20-slim bash -lc 'npm i -g @anthropic-ai\u002Fclaude-code && cd \u002Fworkspace && claude'\n\u002F* No bash Fallback      *\u002F docker run -it --rm -v \"$PWD:\u002Fworkspace\" -e ANTHROPIC_API_KEY=\"sk-your-key\" node:20-slim sh -lc 'npm i -g @anthropic-ai\u002Fclaude-code && cd \u002Fworkspace && claude'\n---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# Check if claude is installed correctly\n\u002F* Linux                 *\u002F which claude\n\u002F* Windows               *\u002F where claude\n\u002F* Universal             *\u002F claude --version\n--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n# Common Management\n\u002F*claude config          *\u002F Configure settings\n\u002F*claude mcp list        *\u002F Setup MCP servers, you can also replace \"list\" with add\u002Fremove\n\u002F*claude agents          *\u002F Open the agent\u002Fsession dashboard\n\u002F*claude update          *\u002F Run a manual update check\n```\n\n---\n\n> [!Tip]\n> \u003Cins>**Open Project Via Terminal Into VS Code \u002F Cursor**\u003C\u002Fins>\n>\n> ### $ - \u003Ckbd>cd \u002Fpath\u002Fto\u002Fproject\u003C\u002Fkbd>\n>\n> ### $ - \u003Ckbd>code .\u003C\u002Fkbd>\n>\n> **Make sure you have the \u003Cmark>(Claude Code extension)\u003C\u002Fmark> installed in your VS Code \u002F Cursor**\n\n---\n\n\u003Ch2 id=\"system-requirements\">System Requirements\u003C\u002Fh2>\n\n> - OS: macOS 10.15+, Ubuntu 20.04+\u002FDebian 10+, or Windows 10\u002F11 or WSL\n\n> - Hardware: 4GB RAM minimum 8GB+ recommended\n\n> - Software: Git 2.23+ is optional for PR\u002Fworktree workflows. Node.js 18+ is only required for npm-based installation; the native installer bundles its own runtime.\n\n> - Internet: Connection for API calls\n\n---\n\n\u003Ch2 id=\"initial-setup\">Initial Setup\u003C\u002Fh2>\n\n> [!Tip]\n> **Find API key from [Anthropic Console](https:\u002F\u002Fconsole.anthropic.com)**\n>\n> **Do NOT commit real keys use git-ignored files, OS key stores, or secret managers**\n\n```C\n# Universal\n\u002F* Authenticate via Anthropic account     *\u002F claude auth login\n\u002F* Authenticate via Console\u002FAPI billing   *\u002F claude auth login --console\n\u002F* Switch accounts inside Claude          *\u002F \u002Flogin\n----------------------------------------------------------------------------------------------------------------------------------\n# Windows\n\u002F* Set-api-key        *\u002F set ANTHROPIC_API_KEY=sk-your-key-here-here\n\u002F* cmd-masked-check   *\u002F echo OK: %ANTHROPIC_API_KEY:~0,8%***\n\u002F* Set-persistent-key *\u002F setx ANTHROPIC_API_KEY \"sk-your-key-here-here\"\n\u002F* cmd-unset-key      *\u002F set ANTHROPIC_API_KEY=\n----------------------------------------------------------------------------------------------------------------------------------\n# Linux\n\u002F* Set-api-key        *\u002F export ANTHROPIC_API_KEY=\"sk-your-key-here-here\"\n\u002F* masked-check       *\u002F echo \"OK: ${ANTHROPIC_API_KEY:0:8}***\"\n\u002F* remove-session     *\u002F unset ANTHROPIC_API_KEY\n----------------------------------------------------------------------------------------------------------------------------------\n# Powershell\n\u002F* ps-session         *\u002F $env:ANTHROPIC_API_KEY = \"sk-your-key-here-here\"\n\u002F* ps-masked-check    *\u002F \"OK: $($env:ANTHROPIC_API_KEY.Substring(0,8))***\"\n\u002F* ps-persist         *\u002F [Environment]::SetEnvironmentVariable(\"ANTHROPIC_API_KEY\",\"sk-your-key-here-here\",\"User\")\n\u002F* ps-rotate          *\u002F $env:ANTHROPIC_API_KEY = \"sk-new-key\"\n\u002F* ps-remove          *\u002F Remove-Item Env:\\ANTHROPIC_API_KEY\n----------------------------------------------------------------------------------------------------------------------------------\n# Other...\n# persist-bash\u002F*      *\u002F echo 'export ANTHROPIC_API_KEY=\"sk-your-key-here-here\"' >> ~\u002F.bashrc && source ~\u002F.bashrc\n# persist-zsh \u002F*      *\u002F echo 'export ANTHROPIC_API_KEY=\"sk-your-key-here-here\"' >> ~\u002F.zshrc  && source ~\u002F.zshrc\n# persist-fish\u002F*      *\u002F fish -lc 'set -Ux ANTHROPIC_API_KEY sk-your-key-here-here'\n----------------------------------------------------------------------------------------------------------------------------------\n```\n\n---\n\n\u003Ch1 id=\"configuration--environment\">Configuration & Environment\u003C\u002Fh1>\n\n\u003Ch2 id=\"environment-variables\">Environment Variables\u003C\u002Fh2>\n\n> **You can also set any of these in settings.json under the \"env\" key for automatic application.**\n\n> [!Important]\n> **Windows Users replace \u003Ckbd>export\u003C\u002Fkbd> with \u003Ckbd>set\u003C\u002Fkbd> or \u003Ckbd>setx\u003C\u002Fkbd> for perm**\n\n```bash\n# Environment toggles (put in ~\u002F.bashrc or ~\u002F.zshrc)\nexport ANTHROPIC_API_KEY=\"sk-your-key-here-here\"      # API key sent as X-Api-Key header (interactive usage: \u002Flogin)\nexport ANTHROPIC_AUTH_TOKEN=\"my-auth-token\"           # Custom Authorization header; Claude adds \"Bearer \" prefix automatically\nexport ANTHROPIC_CUSTOM_HEADERS=\"X-Trace-Id: 12345\"   # Extra request headers (format: \"Name: Value\")\n\nexport ANTHROPIC_MODEL=\"sonnet\"                                  # Custom model name or alias to use\nexport ANTHROPIC_DEFAULT_SONNET_MODEL=\"sonnet\"                   # Default Sonnet alias or pinned Sonnet model ID\nexport ANTHROPIC_DEFAULT_OPUS_MODEL=\"opus\"                       # Default Opus alias or pinned Opus model ID\nexport ANTHROPIC_SMALL_FAST_MODEL=\"haiku-model\"                  # Haiku-class model for background tasks (placeholder)\nexport ANTHROPIC_SMALL_FAST_MODEL_AWS_REGION=\"REGION\"            # Override AWS region for the small\u002Ffast model on Bedrock (placeholder)\n\nexport AWS_BEARER_TOKEN_BEDROCK=\"bedrock_...\"         # Amazon Bedrock API key\u002Ftoken for authentication\n\nexport BASH_DEFAULT_TIMEOUT_MS=60000                  # Default timeout (ms) for long-running bash commands\nexport BASH_MAX_TIMEOUT_MS=300000                     # Maximum timeout (ms) allowed for long-running bash commands\nexport BASH_MAX_OUTPUT_LENGTH=20000                   # Max characters in bash outputs before middle-truncation\n\nexport CLAUDE_BASH_MAINTAIN_PROJECT_WORKING_DIR=1     # (0 or 1) return to original project dir after each Bash command\nexport CLAUDE_BASH_NO_LOGIN=1                         # Force BashTool to skip login shell startup files\nexport CLAUDE_CODE_API_KEY_HELPER_TTL_MS=600000       # Interval (ms) to refresh creds when using apiKeyHelper\nexport CLAUDE_CODE_IDE_SKIP_AUTO_INSTALL=1            # (0 or 1) skip auto-installation of IDE extensions\nexport CLAUDE_CODE_MAX_OUTPUT_TOKENS=4096             # Max number of output tokens for most requests\n\nexport CLAUDE_CODE_USE_BEDROCK=1                      # (0 or 1) use Amazon Bedrock\nexport CLAUDE_CODE_USE_VERTEX=0                       # (0 or 1) use Google Vertex AI\nexport CLAUDE_CODE_SKIP_BEDROCK_AUTH=0                # (0 or 1) skip AWS auth for Bedrock (e.g., via LLM gateway)\nexport CLAUDE_CODE_SKIP_VERTEX_AUTH=0                 # (0 or 1) skip Google auth for Vertex (e.g., via LLM gateway)\n\nexport CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=0     # (0 or 1) disable nonessential traffic (equivalent to DISABLE_* below)\nexport CLAUDE_CODE_DISABLE_TERMINAL_TITLE=0           # (0 or 1) disable automatic terminal title updates\n\nexport CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1         # (0 or 1) enable agent teams research preview\nexport CLAUDE_CODE_ADDITIONAL_DIRECTORIES_CLAUDE_MD=1 # (0 or 1) load CLAUDE.md from --add-dir paths\nexport CLAUDE_CODE_ENABLE_TASKS=false                 # Set to \"false\" to disable the task system\n\nexport DISABLE_AUTOUPDATER=0                          # (0 or 1) disable automatic updates (overrides autoUpdates setting)\nexport DISABLE_BUG_COMMAND=0                          # (0 or 1) disable the \u002Fbug command\nexport DISABLE_COST_WARNINGS=0                        # (0 or 1) disable cost warning messages\nexport DISABLE_ERROR_REPORTING=0                      # (0 or 1) opt out of Sentry error reporting\nexport DISABLE_NON_ESSENTIAL_MODEL_CALLS=0            # (0 or 1) disable model calls for non-critical paths\nexport DISABLE_TELEMETRY=0                            # (0 or 1) opt out of Statsig telemetry\n\nexport HTTP_PROXY=\"http:\u002F\u002Fproxy:8080\"                 # HTTP proxy server URL\nexport HTTPS_PROXY=\"https:\u002F\u002Fproxy:8443\"               # HTTPS proxy server URL\n\nexport MAX_THINKING_TOKENS=0                          # (0 or 1 to turn off\u002Fon) force a thinking budget for the model\nexport MCP_TIMEOUT=120000                             # MCP server startup timeout (ms)\nexport MCP_TOOL_TIMEOUT=60000                         # MCP tool execution timeout (ms)\nexport MAX_MCP_OUTPUT_TOKENS=25000                    # Max tokens allowed in MCP tool responses (default 25000)\n\nexport USE_BUILTIN_RIPGREP=0                          # (0 or 1) set 0 to use system-installed rg instead of bundled one\n\n# Vertex AI region overrides follow VERTEX_REGION_CLAUDE_\u003CMODEL_FAMILY>.\nexport VERTEX_REGION_CLAUDE_3_5_HAIKU=\"REGION\"        # 3.x family example\nexport VERTEX_REGION_CLAUDE_4_6_SONNET=\"REGION\"       # Sonnet family example\nexport VERTEX_REGION_CLAUDE_4_8_OPUS=\"REGION\"         # Opus family example\n\n# -- Session and runtime controls ---------------------------------------------\nexport CLAUDE_CODE_SIMPLE=1                            # Minimal mode: disables MCP tools, attachments, hooks, CLAUDE.md, and skills\nexport CLAUDE_CODE_DISABLE_1M_CONTEXT=1                # Disable 1 M-token context window (use default shorter context)\nexport CLAUDE_CODE_DISABLE_BACKGROUND_TASKS=1          # Disable background tasks entirely\nexport CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS=1        # Opt out of experimental beta features\nexport CLAUDE_CODE_AUTO_CONNECT_IDE=false              # Disable automatic IDE connection on startup\nexport CLAUDE_CODE_TMPDIR=\"\u002Fcustom\u002Ftmp\"                # Override the temp directory Claude Code uses\nexport CLAUDE_CODE_SHELL=\"\u002Fbin\u002Fzsh\"                    # Override shell detection (force a specific shell)\nexport CLAUDE_CODE_SHELL_PREFIX=\"command-wrapper\"      # Prefix\u002Fwrap every shell command Claude runs\nexport CLAUDE_CODE_FILE_READ_MAX_OUTPUT_TOKENS=50000   # Override max output tokens for the Read tool\nexport CLAUDE_CODE_PROXY_RESOLVES_HOSTS=true           # Let the HTTP\u002FHTTPS proxy handle DNS resolution\nexport CLAUDE_CODE_EXIT_AFTER_STOP_DELAY=30000         # Auto-exit SDK mode after idle for N ms\nexport CLAUDE_CODE_PLUGIN_GIT_TIMEOUT_MS=30000         # Timeout (ms) for plugin git operations\nexport CLAUDE_CODE_ACCOUNT_UUID=\"uuid\"                 # Override account UUID (SDK \u002F automation flows)\nexport CLAUDE_CODE_USER_EMAIL=\"user@example.com\"       # Override user email (SDK \u002F automation flows)\nexport CLAUDE_CODE_ORGANIZATION_UUID=\"uuid\"            # Override organization UUID (SDK \u002F automation flows)\nexport ENABLE_CLAUDEAI_MCP_SERVERS=false               # Opt out from claude.ai-synced MCP servers\nexport FORCE_AUTOUPDATE_PLUGINS=1                      # Allow plugin auto-update even when main updater is disabled\nexport IS_DEMO=1                                       # Demo mode — hides email\u002Forg from the UI\nexport NO_PROXY=\"localhost,127.0.0.1\"                  # Bypass proxy for specified hosts (comma-separated)\n\n# -- Provider, terminal, and telemetry controls --------------------------------\nexport CLAUDE_CODE_ENABLE_AUTO_MODE=1                  # Enable auto mode on Bedrock, Vertex, and Foundry for Opus 4.7\u002F4.8\nexport CLAUDE_CODE_USE_POWERSHELL_TOOL=1               # Enable PowerShell tool where available; Windows provider sessions may default to it\nexport CLAUDE_CODE_POWERSHELL_RESPECT_EXECUTION_POLICY=1 # Opt out of PowerShell -ExecutionPolicy Bypass behavior\nexport CLAUDE_CODE_NO_FLICKER=1                        # Prefer flicker-free alternate-screen rendering where supported\nexport CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY=1    # Let compatible gateways populate the model picker from \u002Fv1\u002Fmodels\nexport CLAUDE_CODE_FORCE_SYNC_OUTPUT=1                 # Force synchronized terminal output when auto-detection misses support\nexport CLAUDE_CODE_PACKAGE_MANAGER_AUTO_UPDATE=1       # Let package-manager installs run background upgrades where supported\nexport ENABLE_PROMPT_CACHING_1H=true                   # Opt in to 1-hour prompt cache TTL when your provider supports it\nexport OTEL_LOG_TOOL_DETAILS=1                         # Include tool parameters in tool_decision telemetry events\nexport OTEL_METRICS_INCLUDE_ENTRYPOINT=true            # Include session entrypoint on OpenTelemetry metrics\n```\n\n\u003Ch2 id=\"global-config-options\">Global Config Options\u003C\u002Fh2>\n\n```bash\nclaude config set -g theme dark                               # Theme: dark | light | light-daltonized | dark-daltonized\nclaude config set -g preferredNotifChannel iterm2_with_bell   # Notification channel: iterm2 | iterm2_with_bell | terminal_bell | notifications_disabled\nclaude config set -g autoUpdates true                         # Auto-download & install updates (applied on restart)\nclaude config set -g verbose true                             # Show full bash\u002Fcommand outputs\n\nclaude config set -g attribution false                        # Omit \"co-authored-by Claude\" in git commits\u002FPRs\nclaude config set -g forceLoginMethod claudeai                 # Restrict login flow: claudeai | console\nclaude config set -g model \"sonnet\"                          # Default model override; use a full model ID only when pinning\nclaude config set -g statusLine '{\"type\":\"command\",\"command\":\"~\u002F.claude\u002Fstatusline.sh\"}'  # Custom status line\n\nclaude config set -g enableAllProjectMcpServers true              # Auto-approve all MCP servers from .mcp.json\nclaude config set -g enabledMcpjsonServers '[\"memory\",\"github\"]'  # Approve specific MCP servers\nclaude config set -g disabledMcpjsonServers '[\"filesystem\"]'      # Reject specific MCP servers\n```\n\n\u003Ch2 id=\"configuration-files\">Configuration Files\u003C\u002Fh2>\n\n**(Memory type) Claude Code offers four memory locations in a hierarchical structure, each serving a different purpose:**\n\n| Memory Type                | Location                                                                                                                                                | Purpose                                             | Use Case Examples                                                    | Shared With                     |\n| -------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------- | -------------------------------------------------------------------- | ------------------------------- |\n| **Enterprise policy**      | macOS: `\u002FLibrary\u002FApplication Support\u002FClaudeCode\u002FCLAUDE.md`\u003Cbr \u002F>Linux: `\u002Fetc\u002Fclaude-code\u002FCLAUDE.md`\u003Cbr \u002F>Windows: `C:\\ProgramData\\ClaudeCode\\CLAUDE.md` | Organization-wide instructions managed by IT\u002FDevOps | Company coding standards, security policies, compliance requirements | All users in organization       |\n| **Project memory**         | `.\u002FCLAUDE.md`                                                                                                                                           | Team-shared instructions for the project            | Project architecture, coding standards, common workflows             | Team members via source control |\n| **User memory**            | `~\u002F.claude\u002FCLAUDE.md`                                                                                                                                   | Personal preferences for all projects               | Code styling preferences, personal tooling shortcuts                 | Just you (all projects)         |\n| **Project memory (local)** | `.\u002FCLAUDE.local.md`                                                                                                                                     | Personal project-specific preferences (git-ignored) | Your sandbox URLs, preferred test data, personal overrides           | Just you (current project)      |\n| **Project rules**          | `.claude\u002Frules\u002F*.md`                                                                                                                                    | Modular project rules (loaded alongside CLAUDE.md)  | Linting rules, API conventions, per-directory standards              | Team members via source control |\n\n> All memory files are automatically loaded into Claude Code's context when launched. Files higher in the hierarchy take precedence and are loaded first, providing a foundation that more specific memories build upon.\n\n#### `.claude\u002Frules\u002F` Directory\n\nThe `.claude\u002Frules\u002F` directory lets you break project instructions into separate Markdown files instead of one large `CLAUDE.md`. Every `*.md` file inside is automatically loaded into context alongside `CLAUDE.md`. This is useful for:\n\n- **Modular organization**: Separate concerns (e.g., `api-conventions.md`, `testing-rules.md`)\n- **Per-directory overrides**: Nested `rules\u002F` directories can apply scoped rules\n- **Team collaboration**: Different team members can own different rule files via PR review\n\n#### Auto-Memory\n\nClaude can save useful working context during your sessions, such as project conventions, tooling preferences, and architectural decisions it observes while helping you. Use `\u002Fmemory` to inspect, edit, or remove saved memories.\n\nAuto-memory is most useful for context you would otherwise repeat across sessions:\n\n- Preferred build, test, and lint commands\n- Local conventions that are not obvious from code alone\n- Architecture decisions that influence future edits\n- Team preferences that should shape how Claude proposes changes\n\nKeep durable team rules in `CLAUDE.md` or `.claude\u002Frules\u002F`. Treat auto-memory as helpful working context, not as the only source of truth.\n\n---\n\n\u003Ch1 id=\"commands--usage\">Commands & Usage\u003C\u002Fh1>\n\n\u003Ch2 id=\"claude-commands\">Slash Command Reference\u003C\u002Fh2>\n\n| Command                   | Purpose                                                                                                  |\n| :------------------------ | :------------------------------------------------------------------------------------------------------- |\n| `\u002Fadd-dir`                | Add additional working directories                                                                       |\n| `\u002Fagents`                 | Manage custom AI subagents for specialized tasks                                                         |\n| `\u002Fbranch`                 | Branch or fork the current conversation into a separate session                                          |\n| `\u002Fbug`                    | Report bugs (sends conversation to Anthropic)                                                            |\n| `\u002Fclear`                  | Clear conversation history                                                                               |\n| `\u002Fcompact [instructions]` | Compact conversation with optional focus instructions                                                    |\n| `\u002Fconfig`                 | Open the Settings interface (Config tab)                                                                 |\n| `\u002Fcontext`                | Visualize current context usage as a colored grid                                                        |\n| `\u002Fcopy`                   | Copy conversation content to clipboard                                                                   |\n| `\u002Fcode-review [effort]`   | Run correctness-focused code review; use `--fix` to apply findings or `--comment` for PR comments        |\n| `\u002Fcolor`                  | Set or randomize the current session accent color                                                        |\n| `\u002Fdebug`                  | Troubleshoot current session and diagnose issues                                                         |\n| `\u002Fdoctor`                 | Checks the health of your Claude Code installation                                                       |\n| `\u002Feffort`                 | Pick reasoning effort for the current model\u002Fsession                                                      |\n| `\u002Fexit`                   | Exit the REPL                                                                                            |\n| `\u002Fexport [filename]`      | Export the current conversation to a file or clipboard                                                   |\n| `\u002Ffast`                   | Toggle fast mode for accelerated Opus responses where available                                          |\n| `\u002Fgoal`                   | Set a completion condition so Claude keeps working across turns until it is met                          |\n| `\u002Fhelp`                   | Get usage help                                                                                           |\n| `\u002Finit`                   | Initialize project with CLAUDE.md guide                                                                  |\n| `\u002Finsights`               | Generate an interactive HTML report analyzing your coding habits                                         |\n| `\u002Fkeybindings`            | Configure custom keyboard shortcuts                                                                      |\n| `\u002Flogin`                  | Switch Anthropic accounts                                                                                |\n| `\u002Floop`                   | Schedule a recurring prompt or slash command                                                             |\n| `\u002Flogout`                 | Sign out from your Anthropic account                                                                     |\n| `\u002Fmcp`                    | Manage MCP server connections and OAuth authentication                                                   |\n| `\u002Fmemory`                 | Edit CLAUDE.md memory files                                                                              |\n| `\u002Fmodel`                  | Select or change the AI model                                                                            |\n| `\u002Fpermissions`            | View or update tool permissions                                                                          |\n| `\u002Fplan`                   | Enter plan mode directly from the prompt                                                                 |\n| `\u002Fplugins`                | Manage plugins (install, enable, disable, marketplace)                                                   |\n| `\u002Fpr_comments`            | View pull request comments                                                                               |\n| `\u002Frelease-notes`          | Open the built-in release notes view                                                                     |\n| `\u002Frename \u003Cname>`          | Rename the current session for easier identification                                                     |\n| `\u002Fresume [session]`       | Resume a conversation by ID or name, or open session picker                                              |\n| `\u002Frules`                  | View and manage `.claude\u002Frules\u002F` directory (modular project rules)                                       |\n| `\u002Frewind`                 | Rewind the conversation and\u002For code to a previous point                                                  |\n| `\u002Fsandbox`                | View sandbox dependency status with installation instructions                                            |\n| `\u002Fscroll-speed`           | Tune mouse wheel scroll speed with a live preview                                                        |\n| `\u002Fsettings`               | Open Settings interface (alias for `\u002Fconfig`)                                                            |\n| `\u002Fsimplify`               | Run cleanup-only review for reuse, simplification, efficiency, and altitude                              |\n| `\u002Fstatus`                 | Open Settings interface (Status tab) showing version, model, account                                     |\n| `\u002Fstatusline`             | Set up Claude Code's status line UI                                                                      |\n| `\u002Ftasks`                  | List and manage background tasks                                                                         |\n| `\u002Fteleport`               | Resume a remote session from claude.ai (subscribers only)                                                |\n| `\u002Fterminal-setup`         | Install Shift+Enter key binding for newlines (iTerm2, VS Code, Kitty, Alacritty, Zed, Warp, and WezTerm) |\n| `\u002Fremote-env`             | Configure remote environment settings                                                                    |\n| `\u002Ftheme`                  | Change the color theme                                                                                   |\n| `\u002Ftodos`                  | List current TODO items                                                                                  |\n| `\u002Fultrareview [target]`   | Run comprehensive cloud code review with parallel multi-agent analysis                                   |\n| `\u002Fusage`                  | Show plan usage limits and rate limit status (subscription plans)                                        |\n| `\u002Fusage-credits`          | Enable or inspect usage credits for higher-throughput modes                                              |\n| `\u002Fworkflows`              | View dynamic workflow runs and background orchestration status                                           |\n| `\u002Fbatch`                  | Run batch operations on multiple files (bundled skill)                                                   |\n\n\u003Ch2 id=\"command-line-flags\">Command Line Flags\u003C\u002Fh2>\n\n| Flag \u002F Command                                       | Description                                                                                                                                                  | Example                                                                                |\n| :--------------------------------------------------- | :----------------------------------------------------------------------------------------------------------------------------------------------------------- | :------------------------------------------------------------------------------------- |\n| `-d, --debug`                                        | Enable debug mode (shows detailed debug output).                                                                                                             | `claude -d -p \"query\"`                                                                 |\n| `--include-partial-messages`                         | partial message streaming support via CLI flag                                                                                                               | `claude -p \"query\" --include-partial-messages`                                         |\n| `--verbose`                                          | Override verbose mode setting from config (shows expanded logging \u002F turn-by-turn output).                                                                    | `claude --verbose`                                                                     |\n| `-p, --print`                                        | Print response and exit (useful for piping output).                                                                                                          | `claude -p \"query\"`                                                                    |\n| `--output-format \u003Cformat>`                           | Output format (only works with `--print`): `text` (default), `json` (single result), or `stream-json` (realtime streaming).                                  | `claude -p \"query\" --output-format json`                                               |\n| `--input-format \u003Cformat>`                            | Input format (only works with `--print`): `text` (default) or `stream-json` (realtime streaming input).                                                      | `claude -p --output-format stream-json --input-format stream-json`                     |\n| `--replay-user-messages`                             | Re-emit user messages from stdin back to stdout for acknowledgment — **only works with** `--input-format=stream-json` **and** `--output-format=stream-json`. | `claude --input-format stream-json --output-format stream-json --replay-user-messages` |\n| `--allowedTools`, `--allowed-tools \u003Ctools...>`       | Comma\u002Fspace-separated list of tool names to allow (e.g. `\"Bash(git:*) Edit\"`).                                                                               | `--allowed-tools \"Bash(git:*)\" Edit\"`                                                  |\n| `--disallowedTools`, `--disallowed-tools \u003Ctools...>` | Comma\u002Fspace-separated list of tool names to deny (e.g. `\"Bash(git:*) Edit\"`).                                                                                | `--disallowed-tools \"Edit\"`                                                            |\n| `--mcp-config \u003Cconfigs...>`                          | Load MCP servers from JSON files or strings (space-separated).                                                                                               | `claude --mcp-config .\u002Fmcp-servers.json`                                               |\n| `--strict-mcp-config`                                | Only use MCP servers from `--mcp-config`, ignoring other MCP configurations.                                                                                 | `claude --mcp-config .\u002Fa.json --strict-mcp-config`                                     |\n| `--append-system-prompt \u003Cprompt>`                    | Append a system prompt to the default system prompt (useful in print mode).                                                                                  | `claude -p --append-system-prompt \"Do X then Y\"`                                       |\n| `--permission-mode \u003Cmode>`                           | Permission mode for the session (choices include `acceptEdits`, `auto`, `bypassPermissions`, `default`, `plan`).                                             | `claude --permission-mode plan`                                                        |\n| `--permission-prompt-tool \u003Ctool>`                    | Specify an MCP tool to handle permission prompts in non-interactive mode.                                                                                    | `claude -p --permission-prompt-tool mcp_auth_tool \"query\"`                             |\n| `--fallback-model \u003Cmodel>`                           | Enable automatic fallback to a specified model when the default is overloaded (note: only works with `--print` per help).                                    | `claude -p --fallback-model claude-haiku-20240307 \"query\"`                             |\n| `--model \u003Cmodel>`                                    | Model for the current session. Accepts aliases like `sonnet`\u002F`opus` or a full model ID when pinning.                                                         | `claude --model sonnet`                                                                |\n| `--settings \u003Cfile-or-json>`                          | Load additional settings from a JSON file or a JSON string.                                                                                                  | `claude --settings .\u002Fsettings.json`                                                    |\n| `--add-dir \u003Cdirectories...>`                         | Additional directories to allow tool access to.                                                                                                              | `claude --add-dir ..\u002Fapps ..\u002Flib`                                                      |\n| `--ide`                                              | Automatically connect to an IDE on startup if exactly one valid IDE is available.                                                                            | `claude --ide`                                                                         |\n| `-c, --continue`                                     | Continue the most recent conversation in the current directory.                                                                                              | `claude --continue`                                                                    |\n| `-r, --resume [sessionId]`                           | Resume a conversation; provide a session ID or interactively select one.                                                                                     | `claude -r \"abc123\"`                                                                   |\n| `--session-id \u003Cuuid>`                                | Use a specific session ID for the conversation (must be a valid UUID).                                                                                       | `claude --session-id 123e4567-e89b-12d3-a456-426614174000`                             |\n| `--agents \u003Cjson>`                                    | Define custom subagents dynamically via JSON (see subagent docs for format).                                                                                 | `claude --agents '{\"reviewer\":{\"description\":\"Reviews code\",\"prompt\":\"...\"}}'`         |\n| `--agent \u003Cname>`                                     | Specify a specific agent for the current session.                                                                                                            | `claude --agent my-custom-agent`                                                       |\n| `--bg`                                               | Start or continue work as a background session that can be viewed from `claude agents`.                                                                      | `claude --bg \"fix failing tests\"`                                                      |\n| `--bg --exec \u003Ccommand>`                              | Run a shell command as an attachable background session.                                                                                                     | `claude --bg --exec \"npm test\"`                                                        |\n| `--name \u003Clabel>`                                     | Name a background or remote session for easier identification.                                                                                               | `claude --bg --name nightly-check \"run checks\"`                                       |\n| `--chrome`                                           | Enable Chrome browser integration for web automation and testing.                                                                                            | `claude --chrome`                                                                      |\n| `--no-chrome`                                        | Disable Chrome browser integration for this session.                                                                                                         | `claude --no-chrome`                                                                   |\n| `--remote`                                           | Create a new web session on claude.ai with the provided task description.                                                                                    | `claude --remote \"Fix the login bug\"`                                                  |\n| `--teleport`                                         | Resume a web session in your local terminal.                                                                                                                 | `claude --teleport`                                                                    |\n| `--fork-session`                                     | When resuming, create a new session ID instead of reusing the original.                                                                                      | `claude --resume abc123 --fork-session`                                                |\n| `--json-schema \u003Cschema>`                             | Get validated JSON output matching a JSON Schema after agent completes (print mode only).                                                                    | `claude -p --json-schema '{\"type\":\"object\",...}' \"query\"`                              |\n| `--max-budget-usd \u003Camount>`                          | Maximum dollar amount to spend on API calls before stopping (print mode only).                                                                               | `claude -p --max-budget-usd 5.00 \"query\"`                                              |\n| `--max-turns \u003Cn>`                                    | Limit the number of agentic turns (print mode only). Exits with error when limit reached.                                                                    | `claude -p --max-turns 3 \"query\"`                                                      |\n| `--betas \u003Cheaders>`                                  | Beta headers to include in API requests (API key users only).                                                                                                | `claude --betas interleaved-thinking`                                                  |\n| `--tools \u003Ctools>`                                    | Restrict which built-in tools Claude can use. Use \"\" to disable all, \"default\" for all, or specific tool names.                                              | `claude --tools \"Bash,Edit,Read\"`                                                      |\n| `--system-prompt \u003Cprompt>`                           | Replace the entire system prompt with custom text (works in interactive and print modes).                                                                    | `claude --system-prompt \"You are a Python expert\"`                                     |\n| `--system-prompt-file \u003Cfile>`                        | Load system prompt from a file, replacing the default prompt (print mode only).                                                                              | `claude -p --system-prompt-file .\u002Fcustom-prompt.txt \"query\"`                           |\n| `--append-system-prompt-file \u003Cfile>`                 | Load additional system prompt text from a file and append to default (print mode only).                                                                      | `claude -p --append-system-prompt-file .\u002Fextra-rules.txt \"query\"`                      |\n| `--plugin-dir \u003Cdir>`                                 | Load plugins from directories for this session only (repeatable).                                                                                            | `claude --plugin-dir .\u002Fmy-plugins`                                                     |\n| `--setting-sources \u003Csources>`                        | Comma-separated list of setting sources to load (user, project, local).                                                                                      | `claude --setting-sources user,project`                                                |\n| `--no-session-persistence`                           | Disable session persistence so sessions are not saved to disk (print mode only).                                                                             | `claude -p --no-session-persistence \"query\"`                                           |\n| `--disable-slash-commands`                           | Disable all skills and slash commands for this session.                                                                                                      | `claude --disable-slash-commands`                                                      |\n| `--dangerously-skip-permissions`                     | Bypass all permission checks (only for trusted sandboxes).                                                                                                   | `claude --dangerously-skip-permissions`                                                |\n| `--worktree`, `-w`                                   | Start in an isolated git worktree; `worktree.baseRef` controls whether it branches from fresh remote state or local HEAD.                                    | `claude -w \"implement feature\"`                                                        |\n| `--from-pr \u003Curl>`                                    | Start a session from a pull request URL.                                                                                                                     | `claude --from-pr https:\u002F\u002Fgithub.com\u002Forg\u002Frepo\u002Fpull\u002F123`                                |\n| `--init`                                             | Trigger the Setup hook event.                                                                                                                                | `claude --init`                                                                        |\n| `--init-only`                                        | Run Setup hooks and exit.                                                                                                                                    | `claude --init-only`                                                                   |\n| `--maintenance`                                      | Run Setup hooks in maintenance mode.                                                                                                                         | `claude --maintenance`                                                                 |\n| `-v, --version`                                      | Show the installed `claude` CLI version.                                                                                                                     | `claude --version`                                                                     |\n| `-h, --help`                                         | Display help \u002F usage.                                                                                                                                        | `claude --help`                                                                        |\n\n> The `--output-format json` flag is particularly useful for scripting and automation, allowing you to parse Claude's responses programmatically.\n\n\u003Ch2 id=\"cheat-sheet\">CLI Quick Reference & Configuration Examples\u003C\u002Fh2>\n\n```md\n## Claude Cheat Sheet\n\n# Start and resume\n\nclaude # Start interactive REPL\nclaude \"explain this project\" # Start REPL seeded with a prompt\nclaude -p \"summarize README.md\" # Non-interactive print mode (SDK-backed)\ncat logs.txt | claude -p \"explain\" # Pipe input to Claude and exit\nclaude -c # Continue most recent conversation\nclaude -r \"\u003Csession-id>\" \"finish this\" # Resume by ID or name\nclaude --model sonnet # Pick the Sonnet alias for this run\nclaude --model opus # Pick the Opus alias for harder tasks\n\n# Install, update, and auth\n\nclaude update # Manually update Claude Code\nclaude doctor # Diagnose install\u002Fversion & setup\nclaude install # Start the native binary installer\nclaude migrate-installer # Switch from global npm to the native installer\nclaude auth login # Log in to your Anthropic account\nclaude auth status # Check authentication status\nclaude auth logout # Log out\n\n# Background and remote sessions\n\nclaude agents # Open the live session dashboard: running, blocked, completed\nclaude agents --json # Scriptable JSON list of live\u002Fbackground sessions\nclaude --bg \"run the integration suite and summarize failures\" # Start a background session\nclaude --bg --exec \"npm test\" # Run a shell command as an attachable background session\nclaude remote-control # Start remote-control mode for external tooling\nclaude --remote \"Fix the bug\" # Create web session on claude.ai\nclaude --teleport # Resume web session locally\n\n# Config essentials\n\nclaude config # Interactive config wizard\nclaude config set model \"sonnet\" # Override default model for this project\nclaude config set attribution false # Disable \"co-authored-by Claude\" byline in git\u002FPRs\nclaude config set enableAllProjectMcpServers true # Auto-approve all MCP servers from .mcp.json\nclaude config set defaultMode \"acceptEdits\" # Set default permission mode\nclaude config set worktree.baseRef \"head\" # Use local HEAD instead of origin\u002Fdefault for new worktrees\nclaude config set -g autoUpdates false # Turn off automatic updates globally\nclaude config set -g theme dark # Theme: dark | light | light-daltonized | dark-daltonized\n\n# MCP essentials\n\nclaude mcp # Launch MCP wizard \u002F configure MCP servers\nclaude mcp list # List configured MCP servers\nclaude mcp get \u003Cname> # Show details for a server\nclaude mcp add \u003Cname> \u003Ccommand> [args...] # Add local stdio server\nclaude mcp add --transport http \u003Cname> \u003Curl> # Add remote HTTP server\nclaude mcp reset-project-choices # Reset approvals for project .mcp.json servers\nclaude mcp serve # Run Claude Code itself as an MCP stdio server\n\n# High-value flags\n\nclaude --add-dir ..\u002Fapps ..\u002Flib # Add additional working directories\nclaude --allowedTools \"Bash(git log:\\*)\" \"Read\" # Allow listed tools without permission prompts\nclaude --disallowedTools \"Edit\" # Disallow listed tools without permission prompts\nclaude -p \"query\" --output-format json --input-format stream-json # Control IO formats for scripting\nclaude --verbose # Verbose logging (turn-by-turn)\nclaude --dangerously-skip-permissions # Skip permission prompts (use with caution)\nclaude --permission-mode plan # Start in plan mode (read-only analysis)\nclaude --max-turns 3 -p \"query\" # Limit agentic turns (print mode only)\nclaude --json-schema '{\"type\":\"object\"}' -p \"query\" # Get validated JSON output\nclaude --chrome # Enable Chrome browser integration\nclaude --agent code-reviewer # Run this session with a named agent\nclaude ultrareview 123 --json # Non-interactive comprehensive review for PR\u002Ftarget 123\n\n# Slash shortcuts\n\nclaude --fork-session -r abc123 # Fork instead of reusing original\nclaude -w \"implement feature\" # Start in an isolated git worktree\n\u002Frename auth-refactor # Name current session\n\u002Fresume # Open session picker\n\u002Fexport output.md # Export conversation to file\n\u002Fbranch experiment-name # Branch the current conversation\n\u002Fgoal \"all tests pass and README is updated\" # Keep working until the completion condition is met\n\u002Floop 30m \"check deploy health and summarize anomalies\" # Schedule recurring work\n\u002Fworkflows # View dynamic workflows and background orchestration\n\n# Notes: project scope is default for 'claude config'; use -g\u002F--global for user-global settings.\n# Settings precedence: Enterprise > CLI args > local project > shared project > user (~\u002F.claude).\n```\n\n---\n\n\u003Ch1 id=\"interface--input\">Interface & Input\u003C\u002Fh1>\n\n\u003Ch2 id=\"keyboard-shortcuts\">Keyboard Shortcuts\u003C\u002Fh2>\n\n| Shortcut                     | Description                        | Context                                  |\n| :--------------------------- | :--------------------------------- | :--------------------------------------- |\n| `Ctrl+C`                     | Cancel current input or generation | Standard interrupt                       |\n| `Ctrl+D`                     | Exit Claude Code session           | EOF signal                               |\n| `Ctrl+G`                     | Open in default text editor        | Edit your prompt or custom response      |\n| `Ctrl+L`                     | Clear terminal screen              | Keeps conversation history               |\n| `Ctrl+O`                     | Toggle verbose output              | Shows detailed tool usage and execution  |\n| `Ctrl+R`                     | Reverse search command history     | Search through previous commands         |\n| `Ctrl+V` or `Cmd+V` (iTerm2) | Paste image from clipboard         | Pastes an image or path to an image file |\n| `Ctrl+B`                     | Background running tasks           | Backgrounds bash commands and agents     |\n| `Ctrl+F` (press twice)       | Kill all background agents         | Two-press confirmation to stop agents    |\n| `Up\u002FDown arrows`             | Navigate command history           | Recall previous inputs                   |\n| `Left\u002FRight arrows`          | Cycle through dialog tabs          | Navigate between tabs in dialogs         |\n| `Esc` + `Esc`                | Rewind the code\u002Fconversation       | Restore to a previous point              |\n| `Shift+Tab` or `Alt+M`       | Toggle permission modes            | Switch between Auto-Accept, Plan, Normal |\n| `Option+P` (macOS) \u002F `Alt+P` | Switch model                       | Switch models without clearing prompt    |\n| `Option+T` (macOS) \u002F `Alt+T` | Toggle extended thinking           | Enable\u002Fdisable extended thinking mode    |\n\n\u003Ch3 id=\"text-editing\">Text Editing\u003C\u002Fh3>\n\n| Shortcut               | Description                  | Context                               |\n| :--------------------- | :--------------------------- | :------------------------------------ |\n| `Ctrl+K`               | Delete to end of line        | Stores deleted text for pasting       |\n| `Ctrl+U`               | Delete entire line           | Stores deleted text for pasting       |\n| `Ctrl+Y`               | Paste deleted text           | Paste text deleted with Ctrl+K\u002FU      |\n| `Alt+Y` (after Ctrl+Y) | Cycle paste history          | Cycle through previously deleted text |\n| `Alt+B`                | Move cursor back one word    | Requires Option as Meta on macOS      |\n| `Alt+F`                | Move cursor forward one word | Requires Option as Meta on macOS      |\n\n\u003Ch3 id=\"multiline-input\">Multiline Input\u003C\u002Fh3>\n\n| Method           | Shortcut       | Context                           |\n| :--------------- | :------------- | :-------------------------------- |\n| Quick escape     | `\\` + `Enter`  | Works in all terminals            |\n| macOS default    | `Option+Enter` | Default on macOS                  |\n| Terminal setup   | `Shift+Enter`  | After `\u002Fterminal-setup`           |\n| Control sequence | `Ctrl+J`       | Line feed character for multiline |\n| Paste mode       | Paste directly | For code blocks, logs             |\n\n\u003Ch3 id=\"quick-commands\">Quick Commands\u003C\u002Fh3>\n\n| Shortcut     | Description       | Notes                                 |\n| :----------- | :---------------- | :------------------------------------ |\n| `\u002F` at start | Command or skill  | See built-in commands and skills      |\n| `!` at start | Bash mode         | Run commands directly, add to context |\n| `@`          | File path mention | Trigger file path autocomplete        |\n\n> [!Tip]\n> **PDF Page Ranges:** Use the `pages` parameter with the Read tool for PDFs (e.g., `pages: \"1-5\"`). Large PDFs (>10 pages) return a lightweight reference when @-mentioned instead of being inlined.\n\n\u003Ch2 id=\"vim-mode\">Vim Mode\u003C\u002Fh2>\n\n> [!Note]\n> Enable vim-style editing from `\u002Fconfig` -> Editor mode.\n\n\u003Ch3 id=\"vim-mode-switching\">Vim Mode Switching\u003C\u002Fh3>\n\n| Command | Action                      | From mode |\n| :------ | :-------------------------- | :-------- |\n| `Esc`   | Enter NORMAL mode           | INSERT    |\n| `i`     | Insert before cursor        | NORMAL    |\n| `I`     | Insert at beginning of line | NORMAL    |\n| `a`     | Insert after cursor         | NORMAL    |\n| `A`     | Insert at end of line       | NORMAL    |\n| `o`     | Open line below             | NORMAL    |\n| `O`     | Open line above             | NORMAL    |\n\n\u003Ch3 id=\"vim-navigation\">Vim Navigation\u003C\u002Fh3>\n\n| Command         | Action                    |\n| :-------------- | :------------------------ |\n| `h`\u002F`j`\u002F`k`\u002F`l` | Move left\u002Fdown\u002Fup\u002Fright   |\n| `w`             | Next word                 |\n| `e`             | End of word               |\n| `b`             | Previous word             |\n| `0`             | Beginning of line         |\n| `$`             | End of line               |\n| `^`             | First non-blank character |\n| `gg`            | Beginning of input        |\n| `G`             | End of input              |\n\n\u003Ch3 id=\"vim-editing\">Vim Editing\u003C\u002Fh3>\n\n| Command        | Action                  |\n| :------------- | :---------------------- |\n| `x`            | Delete character        |\n| `dd`           | Delete line             |\n| `D`            | Delete to end of line   |\n| `dw`\u002F`de`\u002F`db` | Delete word\u002Fto end\u002Fback |\n| `cc`           | Change line             |\n| `C`            | Change to end of line   |\n| `cw`\u002F`ce`\u002F`cb` | Change word\u002Fto end\u002Fback |\n| `.`            | Repeat last change      |\n\n> [!Tip]\n> Configure your preferred line break behavior in terminal settings. Run `\u002Fterminal-setup` to install Shift+Enter binding for iTerm2, VS Code, Kitty, Alacritty, Zed, Warp, and WezTerm.\n\n\u003Ch2 id=\"command-history\">Command History\u003C\u002Fh2>\n\n> Claude Code maintains command history for the current session:\n\n```\n* History is stored per working directory\n* Cleared with `\u002Fclear` command\n* Use Up\u002FDown arrows to navigate (see keyboard shortcuts above)\n* **Ctrl+R**: Reverse search through history (if supported by terminal)\n* **Note**: History expansion (`!`) is disabled by default\n```\n\n---\n\n\u003Ch1 id=\"advanced-features\">Advanced Features\u003C\u002Fh1>\n\n\u003Ch2 id=\"thinking-keywords\">Thinking Keywords\u003C\u002Fh2>\n\n> [!Note]\n> **Gives Claude extra pre-answer planning time by adding ONE of these keywords to your prompt.**\n> **Order (lowest → highest) token consumption**\n>\n> \u003Ctable>\u003Ctr>\u003Ctd>\n>\n> > **\u003Ckbd>think\u003C\u002Fkbd> -------------> Lowest**\n>\n> > **\u003Ckbd>think hard\u003C\u002Fkbd>**\n>\n> > **\u003Ckbd>think harder\u003C\u002Fkbd>**\n>\n> > **\u003Ckbd>ultrathink\u003C\u002Fkbd> --------> Highest**\n>\n> \u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftable>\n\n\u003Ch3 id=\"this-makes-claude-spend-more-time\">This makes Claude spend more time:\u003C\u002Fh3>\n\n1. **Planning the solution**\n2. #### breaking down steps\n3. #### weighing alternatives\u002Ftrade-offs\n4. #### checking constraints & edge cases\n   > > #### Higher levels usually increase **latency** and **token usage** pick the smallest that works.\n\n\u003Ch5 id=\"thinking-examples\">Examples\u003C\u002Fh5>\n\n```md\n# Small boost\n\nclaude -p \"Think. Outline a plan to refactor the auth module.\"\n\n# Medium boost\n\nclaude -p \"Think harder. Draft a migration plan from REST to gRPC.\"\n\n# Max boost\n\nclaude -p \"Ultrathink. Propose a step-by-step strategy to fix flaky payment tests and add guardrails.\"\n```\n\n\u003Ch2 id=\"effort-levels\">Effort Levels\u003C\u002Fh2>\n\nUse `\u002Feffort` to tune how much reasoning the selected model applies before answering. Higher effort levels are best for planning-heavy work, deep reviews, and long-context tasks.\n\n```bash\n\u002Feffort            # Open the effort picker\n\u002Feffort low        # Faster, lighter reasoning\n\u002Feffort medium     # Balanced default for many tasks\n\u002Feffort high       # Deeper planning and review\n\u002Feffort xhigh      # Highest effort for Opus 4.8-scale hard tasks\n```\n\nPrefer the lowest effort that still solves the task: higher effort can improve planning, code review, and long-context reasoning, but it usually increases latency and token usage.\n\n\u003Ch2 id=\"fast-mode\">Fast Mode\u003C\u002Fh2>\n\n> [!Note]\n> **Fast Mode provides accelerated Opus responses for rapid iteration when speed matters more than maximum depth.**\n\n**How to enable Fast Mode:**\n\n```bash\n# Enable usage credits if your plan requires it, then toggle fast mode\n\u002Fusage-credits\n\u002Ffast\n\n# Or toggle during conversation\n# The status bar will show when Fast Mode is active\n```\n\n**Key features:**\n\n- **Faster responses** - Reduced latency for quick tasks\n- **Opus support** - Use with Opus models where fast mode is available\n- **Usage credits** - Some plans require `\u002Fusage-credits` before `\u002Ffast`\n- **Visible state** - The status bar and IDE indicators show when Fast Mode is active\n\n**When to use Fast Mode:**\n\n- Quick code reviews and edits\n- Rapid prototyping\n- Simple questions and commands\n- Iterative debugging\n\n> Fast Mode trades some depth for speed. Use normal mode for complex analysis and planning tasks.\n\n\u003Ch2 id=\"auto-mode\">Auto Mode\u003C\u002Fh2>\n\nAuto mode lets Claude evaluate and approve lower-risk actions automatically while still blocking or asking on higher-risk operations. It is useful for trusted development loops where repeated permission prompts slow down work.\n\n```bash\n# Enable auto mode for Bedrock, Vertex, and Foundry Opus 4.7\u002F4.8 sessions\nexport C","2026-06-11 04:11:17","high_star"]