Mastering Claude Code CLI 🚀

Navigate the terminal like a pro with these essential commands.

Claude Code is a terminal-based agent that works inside your project.

What is Claude Code?

Claude Code is Anthropic's agentic coding assistant that runs in your terminal. Unlike web-based AI tools, it can:

✅ Executes real terminal commands rather than just suggesting them
✅ Reads and edits files directly in your project
✅ Manages git operations (diff, commit, PR comments)
✅ Remembers project context across sessions
✅ Integrates with MCP servers for more capabilities

💡 Think of it as pair programming with an AI that has direct access to your codebase.

Session & conversation control

These commands manage your sessions and context window.

Command	What It Does
`/clear`	Resets conversation history to free up context window
`/compact`	Summarizes the conversation to save space while keeping focus
`/rewind`	Returns the conversation or code to a specific previous checkpoint
`/resume`	Opens the session picker to continue a previous discussion
`/rename`	Gives the current session a custom name for easier tracking
`/exit`	Safely closes the CLI session and returns to your shell

Core development workflow

Commands for planning your work and reviewing it before you commit.

Command	What It Does
`/plan`	Enters a dedicated mode to outline tasks before execution
`/diff`	Opens an interactive viewer to inspect uncommitted changes
`/add-dir`	Grants Claude access to specific directories for file operations
`/init`	Initializes a project with CLAUDE.md and core configurations
`/copy`	Copies specific responses or code blocks to your clipboard
`/security-review`	Analyzes pending code changes for potential vulnerabilities

Model & performance tuning

Adjust the model, reasoning depth, and cost to fit the task.

Command	What It Does
`/model`	Switches between different AI models or adjusts effort levels
`/effort`	Sets the reasoning depth (low to max) for complex tasks
`/fast`	Toggles fast mode to prioritize speed over deep reasoning
`/context`	Visualizes current token usage and optimization suggestions
`/cost`	Displays detailed token usage and subscription statistics
`/usage`	Shows your remaining plan limits and rate limit status

Interface & personalization

Make the CLI feel the way you want.

Command	What It Does
`/config`	Opens the settings UI for themes and output preferences
`/theme`	Changes UI colors (Light, Dark, or System palettes)
`/color`	Sets a specific highlight color for the prompt bar
`/vim`	Toggles between Vim-style and Normal text editing modes
`/keybindings`	Configures custom shortcuts for terminal operations
`/statusline`	Customizes the information displayed at the bottom of the CLI

System & integration

Manage connections, permissions, and account access.

Command	What It Does
`/mcp`	Manages Model Context Protocol (MCP) server connections
`/permissions`	Manages "allow" or "deny" rules for automated tool use
`/doctor`	Diagnoses and verifies your installation and connectivity
`/login`	Authenticates the CLI with your Anthropic account
`/pr-comments`	Fetches and displays comments from GitHub Pull Requests
`/memory`	Edits persistent project memory and auto-memory files

Pro tip & summary

Combine /plan for architecture, /diff for review, and /security-review before committing. That sequence keeps AI-generated code intentional and secure.

Ready to level up? Type /help in your terminal to explore all 30+ available commands.

Hashtags

#ClaudeCode #Anthropic #CLI #Productivity #CodingTools #DeveloperTools #AI #Terminal #Git #AgenticAI

RubixScript Blog

🤖 AI Agent Development in 2025: From Zero to Autonomous Systems