claude-code-command-mastery - SKILL.md Agent Skill

name: claude-code-command-mastery description: Master essential Claude Code terminal commands including slash commands, agent creation, skills management, security reviews, ultra think mode, plan mode, and IDE integration. Trigger when setting up Claude Code workflows, optimizing development processes, or needing comprehensive command reference. Apply for terminal-based AI development, code security, and workflow automation. source: 10 Claude Code tips I wish I knew from the start + https://youtu.be/BEIulrjHzMI?si=JiZ_XdfyLiw9nCGP generated_by: SkillJack (skilljacked.com)

Claude Code Command Mastery

Master the essential terminal commands and features for effective Claude Code development workflows.

Quick Start

# In Claude Code terminal
/  # Shows all available commands
/compact  # Clear conversation history but keep summary
/security  # Review pending changes for security issues

Core Workflow

When starting any Claude Code session:

Type / to see all available commands
Use /compact [optional instructions] when chat gets too long
Enable plan mode with Shift+Tab before complex builds
Create claude.md file in project root for persistent context

When building complex applications:

Use /plan or Shift+Tab to enable planning mode
Let Claude ask clarifying questions before coding
Accept the plan only when satisfied with approach

When needing specialized help:

Use /agents to create or access specialized sub-agents
Use /skills to apply reusable capabilities
Stack multiple skills for complex workflows

Techniques

Slash Commands (Essential Set)

/compact [instructions] — Compress conversation history with optional focus
/security — Security review of pending changes on current branch
/agents — Create/manage specialized sub-agents with own context
/skills — List/create/apply reusable capabilities
/plan — Enable planning mode (alternative: Shift+Tab)

Agent Creation Pattern

/agents
→ Create new agent
→ Describe role: "marketer who can help me write amazing brand copy"
→ Choose tools (read-only/full access)
→ Select model (Sonnet/Opus/Haiku)
→ Assign color
→ Save

Agent Invocation

"Use the agent [agent-name] to [specific task]"
# Example: "Use the agent brand copywriter to generate five headline ideas for a coffee shop website"

Skills vs Agents

Skills: Work within current context window, stackable, reusable across sessions
Agents: Own context window, own system prompt, work independently like employees

Ultra Think Mode

Type ultra think before any prompt to make Claude use more tokens for deeper reasoning (shows rainbow color indicator).

IDE Integration Setup

Open Claude Code terminal inside IDE (Cursor, VS Code, etc.)
Navigate to project folder
Claude automatically accesses all files in folder
Use both IDE features and Claude Code together

Claude.md Configuration

Create claude.md in project root with:

# Custom Instructions
- Always type-check after code changes
- Follow [specific coding style]
- Use [preferred frameworks]
- [Workflow preferences]

Anti-Patterns

NEVER:

Skip planning mode for complex applications — leads to unfocused development
Let conversation history grow indefinitely — wastes tokens and context
Create agents for simple, one-time tasks — use direct prompts instead
Stack too many skills simultaneously — can create conflicting instructions
Ignore the claude.md file — missing persistent context hurts consistency

Edge Cases & Error Handling

Long conversation degradation:

Use /compact when responses become less focused
Include summarization instructions for domain-specific focus

Agent context confusion:

Agents work in separate context windows — they don't see main conversation
Always provide full context when invoking agents

Skill conflicts:

When stacking skills, ensure they complement rather than contradict
Test skill combinations on simple tasks first

Plan mode confusion:

Plan mode prevents immediate code execution — accept plan to proceed
Use Shift+Tab to toggle plan mode on/off

Bundled Resources Plan

scripts/agent-templates.md — Pre-configured agent templates for common roles
scripts/skill-library.md — Curated skills for different development domains
references/claude-md-examples.md — Sample claude.md configurations
references/slash-command-reference.md — Complete slash command documentation