/plan
Helps you plan before coding. Creates detailed strategies to make development smoother.
> /plugin marketplace add wasabeef/claude-code-cookbookHow it fires
How this command gets triggered: by you, by Claude, or both.
- Fires itselfClaude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/plan
Context preview
What this command does when you run it.
Helps you plan before coding. Creates detailed strategies to make development smoother.
Command definition
plan.mdPlan mode for implementation strategy before coding
Helps you plan before coding. Creates detailed strategies to make development smoother.
Usage
# Request Plan Mode from Claude
"Create an implementation plan for [implementation content]"
Basic Examples
# Implementation plan for new feature
"Create an implementation plan for user authentication functionality"
# System design plan
"Create an implementation plan for microservice splitting"
# Refactoring plan
"Create a refactoring plan for legacy code"
Collaboration with Claude
# Complex feature implementation
"Create an implementation plan for chat functionality, including WebSocket, real-time notifications, and history management"
# Database design
"Create a database design plan for an e-commerce site, including product, order, and user management"
# API design
"Create an implementation plan for GraphQL API, including authentication, caching, and rate limiting"
# Infrastructure design
"Create an implementation plan for Dockerization, including development environment, production environment, and CI/CD"
How Plan Mode Works
**Automatic Start**
- Starts automatically when you describe what to build
- Or just say "Create an implementation plan"
**What You Get**
- Clear requirements (user stories, success criteria)
- Design docs (architecture, data model, UI)
- Implementation steps (tasks, tracking, quality checks)
- Risk analysis and solutions
**Getting Your Approval**
- I'll show you the plan using `ExitPlanMode`
- **Important**: I always wait for your explicit OK
- I won't code without your approval
- You can request changes anytime
- TaskCreate tracking starts after you approve
Detailed Examples
# Complex system implementation
"Create an implementation plan for an online payment system, including Stripe integration, security, and error handling"
# Frontend implementation
"Create an implementation plan for a React dashboard, including state management, component design, and testing"
# Backend implementation
"Create an implementation plan for a RESTful API, including authentication, validation, and logging"
# DevOps implementation
"Create an implementation plan for a CI/CD pipeline, including test automation, deployment, and monitoring"
3-Phase Workflow
Phase 1: Requirements
- **User Stories**: What are we building and why?
- **Success Criteria**: How do we know it's done?
- **Constraints**: What limits do we have?
- **Priority**: What's must-have vs nice-to-have?
Phase 2: Design
- **Architecture**: How will the system work?
- **Data Model**: Database schema and APIs
- **UI/UX**: Screen layouts and user flow
- **Risks**: What could go wrong and how to prevent it
Phase 3: Implementation
- **Task Breakdown**: Split into manageable chunks
- **Progress Tracking**: TaskUpdate manages status
- **Quality Checks**: Testing and verification plan
- **Your Approval**: Show plan and wait for your OK
Notes
**When to Use This**
- Best for complex projects
- Skip for simple fixes
- Great for 3+ step tasks or new features
**Technical Notes**
- Don't rely on `ExitPlanMode` return values
- Only your explicit approval counts
- Works differently than CLI plan mode
**Important Rules**
- Never start coding before you approve
- Always wait for your response
- Offer alternatives if something fails
Execution Example
# Usage example
"Create an implementation plan for a user management system"
# What happens:
# 1. Plan Mode starts
# 2. Analyze requirements and pick tech
# 3. Structure the implementation
# 4. Show you the plan
# 5. Start coding after you approve
Read more
Plan mode for implementation strategy before coding
Helps you plan before coding. Creates detailed strategies to make development smoother.
Usage
# Request Plan Mode from Claude "Create an implementation plan for [implementation content]"
Basic Examples
# Implementation plan for new feature "Create an implementation plan for user authentication functionality" # System design plan "Create an implementation plan for microservice splitting" # Refactoring plan "Create a refactoring plan for legacy code"
Collaboration with Claude
# Complex feature implementation "Create an implementation plan for chat functionality, including WebSocket, real-time notifications, and history management" # Database design "Create a database design plan for an e-commerce site, including product, order, and user management" # API design "Create an implementation plan for GraphQL API, including authentication, caching, and rate limiting" # Infrastructure design "Create an implementation plan for Dockerization, including development environment, production environment, and CI/CD"
How Plan Mode Works
**Automatic Start**
- Starts automatically when you describe what to build
- Or just say "Create an implementation plan"
**What You Get**
- Clear requirements (user stories, success criteria)
- Design docs (architecture, data model, UI)
- Implementation steps (tasks, tracking, quality checks)
- Risk analysis and solutions
**Getting Your Approval**
- I'll show you the plan using `ExitPlanMode`
- **Important**: I always wait for your explicit OK
- I won't code without your approval
- You can request changes anytime
- TaskCreate tracking starts after you approve
Detailed Examples
# Complex system implementation "Create an implementation plan for an online payment system, including Stripe integration, security, and error handling" # Frontend implementation "Create an implementation plan for a React dashboard, including state management, component design, and testing" # Backend implementation "Create an implementation plan for a RESTful API, including authentication, validation, and logging" # DevOps implementation "Create an implementation plan for a CI/CD pipeline, including test automation, deployment, and monitoring"
3-Phase Workflow
Phase 1: Requirements
- **User Stories**: What are we building and why?
- **Success Criteria**: How do we know it's done?
- **Constraints**: What limits do we have?
- **Priority**: What's must-have vs nice-to-have?
Phase 2: Design
- **Architecture**: How will the system work?
- **Data Model**: Database schema and APIs
- **UI/UX**: Screen layouts and user flow
- **Risks**: What could go wrong and how to prevent it
Phase 3: Implementation
- **Task Breakdown**: Split into manageable chunks
- **Progress Tracking**: TaskUpdate manages status
- **Quality Checks**: Testing and verification plan
- **Your Approval**: Show plan and wait for your OK
Notes
**When to Use This**
- Best for complex projects
- Skip for simple fixes
- Great for 3+ step tasks or new features
**Technical Notes**
- Don't rely on `ExitPlanMode` return values
- Only your explicit approval counts
- Works differently than CLI plan mode
**Important Rules**
- Never start coding before you approve
- Always wait for your response
- Offer alternatives if something fails
Execution Example
# Usage example "Create an implementation plan for a user management system" # What happens: # 1. Plan Mode starts # 2. Analyze requirements and pick tech # 3. Structure the implementation # 4. Show you the plan # 5. Start coding after you approve
A collection of commands, roles, and automation scripts for Claude Code. Automate your workflow without unnecessary confirmations, allowing you to focus on what matters.
Repo: wasabeef/claude-code-cookbook
Other commands on claude-code-cookbook.
- /analyze-dependencies
Analyzes your project's dependencies and checks architecture health.
Open command - /analyze-performance
Analyzes application performance from a user experience perspective and quantifies experience improvements from optimizations. Calculates UX scores based on Core Web Vitals and proposes prioritized optimization strategies.
Open command - /check-fact
Verifies if a statement is true by checking your project's code and documentation.
Open command - /check-prompt
A comprehensive collection of best practices for evaluating and improving the quality of prompts for AI Agents. It systematizes knowledge gained from actual prompt improvement processes, covering all important aspects such as ambiguity elimination, information integration,
Open command - /commit-message
Generates commit messages from staged changes (git diff --staged). This command only creates messages and copies them to your clipboardβit doesn't run any git commands.
Open command - /context7
Searches technical documentation using MCP's Context7.
Open command

