Cheatsheet
85 slash commands - the ones that change your workflow
POPULARRead time: 5 min
title: "Cheatsheet" description: "85 slash commands - the ones that change your workflow" section: "Getting Started" readTime: "5 min" badge: "POPULAR"
Claude Code Cheatsheet
85 slash commands. Most people know 5. Here are the ones that change your workflow.
Essential Commands
| Command | What It Does |
|---|---|
/plan | Preview changes before execution (Plan Mode) |
/compact | Compress session history (70% threshold) |
/clear | Clear session context (90% threshold) |
/context | Show current context (files, tools, rules) |
/cost | Display token usage and costs |
/stats | Session statistics |
Bundled Skills
Pre-loaded skill commands:
| Command | Purpose | Example |
|---|---|---|
/batch | Run parallel agents | /batch "Task 1" "Task 2" "Task 3" |
/simplify | 3-agent code review | /simplify on complex code |
/debug | Interactive debugger | /debug, then paste error |
/loop | Agentic loop for complex tasks | /loop "Implement X with tests" |
/claude-api | Load API reference | /claude-api (in-session docs) |
Harness Commands (ECC-Powered)
| Command | Purpose |
|---|---|
/harness-audit | Reliability + eval readiness audit |
/loop-start | Start controlled agentic loop |
/loop-status | Check loop execution status |
/quality-gate | Run checks on paths/repo |
/model-route | Route tasks by complexity/budget |
Permission Modes
Cycle with Shift+Tab:
- default - Ask before each change
- acceptEdits - Accept file changes, ask about command execution
- plan - Show plan without executing
- auto - Classify actions and decide automatically
- dontAsk - Never ask, log decisions
- bypassPermissions - Unrestricted (not recommended)
Set with:
claude --permission-mode auto "your task"Session Management
| Flag | Purpose | Example |
|---|---|---|
--resume | Resume previous session | claude --resume |
--fork-session | Branch session | claude --fork-session SESSION_ID |
--from-pr | Load PR context | claude --from-pr 42 "review changes" |
--effort | Task difficulty hint | claude --effort high "complex refactor" |
Compaction Strategy
Token Usage
50% → Monitor
70% → /compact
90% → /clear or new session
What /compact preserves:
- File edits
- Task completion status
- Key decisions
- Tool results
What it removes:
- Verbose reasoning
- Intermediate steps
- Duplicate context
Anti-Patterns to Avoid
❌ Task without recent CLAUDE.md update
❌ Task without clear scope
❌ Using /clear mid-implementation
❌ Running agent teams without checkpoints
❌ Bypassing permissions in complex work
Pro Tip: Combine /plan with --effort low for quick exploratory runs, then /batch for parallel implementation with --effort high.