Plan2Code: AI-Assisted Software Development Workflow

A structured 4-step workflow for developing features and projects with AI assistance. This methodology emphasizes thorough planning before implementation, ensuring well-documented, maintainable code.

Overview

        🤔                     📝                     ⚡                     🧹
┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│   Step 1        │     │   Step 2        │     │   Step 3        │     │   Step 4        │
│   PLAN          │ --> │   DOCUMENT      │ --> │   IMPLEMENT     │ --> │   FINALIZE      │
└─────────────────┘     └─────────────────┘     └─────────────────┘     └─────────────────┘
      New Chat               New Chat           New Chat (per phase)        New Chat

Step	File	Purpose
1	plan2code-1--plan.md	Requirements analysis and architecture design
2	plan2code-2--document.md	Create structured implementation documentation
3	plan2code-3--implement.md	Execute the implementation phase by phase
4	plan2code-4--finalize.md	Validate, summarize, and archive

---

Installation

This repository includes pre-configured workflow files for all major AI coding assistants. Choose your platform and follow the setup instructions below.

Supported Platforms

Platform	Directory	Invocation	Status
Claude Code CLI	.claude/commands/	/plan2code-1--plan, /plan2code-2--document, /plan2code-3--implement, /plan2code-4--finalize	Ready
GitHub Copilot CLI	.github/agents/	--agent=plan2code-1--plan or /agent plan2code-1--plan	Ready
VS Code Copilot	.github/prompts/	Slash commands in chat	Ready
Windsurf IDE	.windsurf/workflows/	/plan2code-1--plan, /plan2code-2--document, /plan2code-3--implement, /plan2code-4--finalize	Ready
Cursor AI	.cursor/rules/	Command palette or auto-apply	Ready
Google Antigravity	.agent/workflows/	/plan2code-1--plan, /plan2code-2--document, /plan2code-3--implement, /plan2code-4--finalize	Ready
Continue	.continue/prompts/	/plan2code-1--plan, /plan2code-2--document, /plan2code-3--implement, /plan2code-4--finalize	Ready

Quick Start

Option A: Clone and copy to your project

# Clone the repository
git clone https://github.com/your-username/plan2code.git

# Copy the platform-specific directory to your project
# Example for Claude Code:
cp -r plan2code/.claude your-project/

# Example for Cursor:
cp -r plan2code/.cursor your-project/

Option B: Copy individual platform directories

Download only the directories you need for your AI coding tool.

---

Claude Code CLI

Location: .claude/commands/

Setup:

1. Copy the .claude/commands/ directory to your project root
2. Restart Claude Code or start a new session

Files:

.claude/commands/
├── plan2code-1--plan.md       # Step 1: Planning
├── plan2code-2--document.md   # Step 2: Documentation
├── plan2code-3--implement.md  # Step 3: Implementation
└── plan2code-4--finalize.md   # Step 4: Finalization

Usage:

/plan2code-1--plan           # Start planning a new feature
/plan2code-2--document       # Create implementation docs from plan
/plan2code-3--implement      # Begin/continue implementation
/plan2code-4--finalize       # Wrap up after all phases complete

Documentation: Claude Code Slash Commands

---

GitHub Copilot CLI

Location: .github/agents/

Setup:

1. Copy the .github/agents/ directory to your project root
2. Ensure GitHub Copilot CLI is installed: npm install -g @github/copilot@latest

Files:

.github/agents/
├── plan2code-1--plan.agent.md       # Step 1: Planning
├── plan2code-2--document.agent.md   # Step 2: Documentation
├── plan2code-3--implement.agent.md  # Step 3: Implementation
└── plan2code-4--finalize.agent.md   # Step 4: Finalization

Usage:

# Using --agent flag
copilot --agent=plan2code-1--plan --prompt "I want to build a REST API"

# Using slash commands in interactive mode
copilot
> /agent plan2code-1--plan

Documentation: GitHub Copilot CLI Custom Agents

---

VS Code GitHub Copilot

Location: .github/prompts/

Setup:

1. Copy the .github/prompts/ directory to your project root
2. Open VS Code and ensure GitHub Copilot extension is installed
3. Prompts are automatically recognized

Files:

.github/prompts/
├── plan2code-1--plan.prompt.md       # Step 1: Planning
├── plan2code-2--document.prompt.md   # Step 2: Documentation
├── plan2code-3--implement.prompt.md  # Step 3: Implementation
└── plan2code-4--finalize.prompt.md   # Step 4: Finalization

File Format:

---
mode: agent
description: "Plan2Code Step 1: Planning Mode"
---
[prompt content]

Usage:

• Open Copilot Chat (Ctrl+Shift+I or Cmd+Shift+I)
• Type / to see available prompts
• Select the desired workflow step

Documentation: VS Code Copilot Prompt Files

---

Windsurf IDE

Location: .windsurf/workflows/

Setup:

1. Copy the .windsurf/workflows/ directory to your project root
2. Workflows appear automatically in Cascade

Files:

.windsurf/workflows/
├── plan2code-1--plan.md       # Step 1: Planning
├── plan2code-2--document.md   # Step 2: Documentation
├── plan2code-3--implement.md  # Step 3: Implementation
└── plan2code-4--finalize.md   # Step 4: Finalization

Note: Windsurf has a 12,000 character limit per workflow file.

Usage:

• In Cascade, type /plan2code-1--plan to invoke the planning workflow

Documentation: Windsurf Workflows

---

Cursor AI

Location: .cursor/rules/

Setup:

1. Copy the .cursor/rules/ directory to your project root
2. Rules are recognized automatically

Files:

.cursor/rules/
├── plan2code-1--plan.mdc       # Step 1: Planning
├── plan2code-2--document.mdc   # Step 2: Documentation
├── plan2code-3--implement.mdc  # Step 3: Implementation
└── plan2code-4--finalize.mdc   # Step 4: Finalization

File Format (MDC):

---
description: "Plan2Code Step 1: Planning Mode"
alwaysApply: false
---
[prompt content]

Usage:

• Open Command Palette (Cmd+Shift+P)
• Type "New Cursor Rule" to create new rules
• Reference rules by mentioning them in chat
• Use /Generate Cursor Rules to create rules from conversation

Documentation: Cursor Rules for AI

---

Google Antigravity

Location: .agent/workflows/

Setup:

1. Copy the .agent/workflows/ directory to your project root
2. Workflows appear automatically in Antigravity

Files:

.agent/workflows/
├── plan2code-1--plan.md       # Step 1: Planning
├── plan2code-2--document.md   # Step 2: Documentation
├── plan2code-3--implement.md  # Step 3: Implementation
└── plan2code-4--finalize.md   # Step 4: Finalization

File Format:

---
description: "Plan2Code Step 1: Planning Mode"
---
[prompt content]

Usage:

• Type /plan2code-1--plan in the agent chat to invoke the planning workflow

Documentation: Customize Antigravity

---

Continue (VS Code/JetBrains)

Location: .continue/prompts/

Setup:

1. Copy the .continue/prompts/ directory to your project root
2. Install the Continue extension for VS Code or JetBrains
3. Prompts are automatically recognized

Files:

.continue/prompts/
├── plan2code-1--plan.prompt.md       # Step 1: Planning
├── plan2code-2--document.prompt.md   # Step 2: Documentation
├── plan2code-3--implement.prompt.md  # Step 3: Implementation
└── plan2code-4--finalize.prompt.md   # Step 4: Finalization

File Format:

---
name: plan2code-1--plan
description: "Plan2Code Step 1: Planning Mode"
---
[prompt content]

Usage:

• In Continue chat, type /plan2code-1--plan to invoke the planning workflow
• Use {{{ input }}} template variable for user input
• Use {{{ currentFile }}} to reference the current file

Documentation: Continue Prompts

---

Manual Installation (Any Platform)

If your AI tool isn't listed above, you can still use Plan2Code:

1. Copy/Paste Method: Copy the contents of the appropriate plan2code-*.md file and paste it at the start of your conversation.

2. File Reference Method: Reference the file directly in your prompt or use the @plan2code-1--plan.md syntax:

   Please follow the instructions in plan2code-1--plan.md
   
   I want to build a user authentication system.
   

3. Custom Integration: Adapt the prompts to your tool's custom instruction format.

---

Important: Start Fresh Conversations

Start a new conversation/chat session before each step. This includes:

• Step 1: New conversation
• Step 2: New conversation
• Step 3: New conversation for each phase (Phase 1, Phase 2, etc.)
• Step 4: New conversation

Fresh conversations prevent context pollution and ensure the AI focuses on the current task with the relevant specifications.

---

The Workflow Steps

Step 1: Planning Mode 🤔

Purpose: Thoroughly analyze requirements and design the solution architecture before writing any code.

AI Role: Senior software architect and technical product manager

Phases (completed one at a time):

1. Requirements Analysis - Extract functional/non-functional requirements, identify ambiguities
2. System Context Examination - Review existing codebase, identify integration points
3. Tech Stack - Recommend and confirm all technologies (requires user sign-off)
4. Architecture Design - Propose patterns, define components, design interfaces/schemas
5. Technical Specification - Break down implementation phases, identify risks
6. Transition Decision - Finalize plan when confidence reaches 90%+

Output: specs/PLAN-DRAFT-<timestamp>.md containing the complete implementation plan

Key Behaviors:

• AI stops after each phase for clarification
• Must reach 90% confidence before finalizing
• All assumptions are documented
• User must approve tech stack decisions

---

Step 2: Documentation Mode 📝

Purpose: Transform the planning output into structured, actionable implementation documents.

Required Context: Attach or reference the specs/PLAN-DRAFT-<timestamp>.md from Step 1 (or provide the planning conversation).

Output Structure:

specs/
└── <feature-name>/
    ├── overview.md          # High-level overview with phase checkboxes
    ├── Phase 1.md           # Detailed tasks for Phase 1
    ├── Phase 2.md           # Detailed tasks for Phase 2
    └── Phase N.md           # ...additional phases

Document Format:

• Each phase file contains detailed one-story-point tasks
• All tasks have checkboxes [ ] for progress tracking
• Each phase is self-contained (developer needs no prior context)
• Unit/E2E testing excluded unless explicitly requested

---

Step 3: Implementation Mode ⚡

Purpose: Execute the implementation following the documented specifications.

AI Role: Senior software engineer

Required Context: Attach or reference the specs/<feature-name>/ directory contents:

• overview.md (to identify which phase is next)
• The relevant Phase X.md file(s)

Workflow:

1. Identify the next uncompleted phase (unchecked in overview.md)
2. Implement ALL tasks in that phase exactly as specified
3. Update Phase X.md checkboxes as tasks complete [x]
4. Update overview.md phase checkbox when phase completes
5. Perform code review to ensure nothing was missed
6. Add completion summary to the phase document

Key Rules:

• Start a new conversation for EACH phase
• Work on ONE phase per conversation (unless told otherwise)
• Follow specifications EXACTLY as documented
• Keep checkboxes updated (enables progress tracking across sessions)
• Do NOT run tests unless specified in phase tasks

---

Step 4: Finalization Mode 🧹

Purpose: Validate implementation, create summaries, and archive documentation.

Required Context: Attach or reference the specs/<feature-name>/ directory contents.

Steps:

1. Validation - Verify all tasks implemented correctly, check for issues
2. Summary - Document what was built and list all modified/created files
3. Documentation Review - Identify any needed README/CHANGELOG updates
4. Spec Cleanup - Move completed specs to specs/completed/<implementation-name>/
5. Final Confirmation - Confirm completion

---

How to Use These Prompts

Option 1: Platform-Specific Slash Commands (Recommended)

This repository includes pre-configured workflow files for all major AI coding assistants. See the Installation section above for platform-specific setup instructions.

Quick commands after setup:

/plan           # Start planning a new feature
/document       # Create implementation docs from plan
/implement      # Begin/continue implementation
/finalize       # Wrap up after all phases complete

Option 2: Direct File Reference

Reference the prompt files directly in your conversation:

Please follow the instructions in plan2code-1--plan.md

I want to build a user authentication system with OAuth support.

Option 3: Copy/Paste

Copy the contents of each prompt file and paste at the beginning of your conversation when starting that step.

---

Complete Workflow Example

Starting a New Project

Session 1 - Planning (New Chat):

User: [Paste or invoke Step 1 prompt]
      I want to build a REST API for a task management application.

AI:   🤔 [REQUIREMENTS ANALYSIS]
      ... asks clarifying questions, works through phases ...

AI:   🤔 [TRANSITION DECISION]
      Confidence: 92%. Creating specs/PLAN DRAFT.md...

Session 2 - Documentation (New Chat):

User: [Paste or invoke Step 2 prompt]
      [Attach: specs/PLAN DRAFT.md]

AI:   📝 [DOCUMENTATION]
      Creating specs/task-api/overview.md...
      Creating specs/task-api/Phase 1.md...
      Creating specs/task-api/Phase 2.md...
      ...

Session 3 - Implementation Phase 1 (New Chat):

User: [Paste or invoke Step 3 prompt]
      [Attach: specs/task-api/overview.md]
      [Attach: specs/task-api/Phase 1.md]

AI:   ⚡ [PHASE 1: Project Setup]
      Implementing tasks...
      ✓ Phase 1 complete. Updated checkboxes in Phase 1.md and overview.md.

Session 4 - Implementation Phase 2 (New Chat):

User: [Paste or invoke Step 3 prompt]
      [Attach: specs/task-api/overview.md]
      [Attach: specs/task-api/Phase 2.md]

AI:   ⚡ [PHASE 2: Database Models]
      Implementing tasks...
      ✓ Phase 2 complete. Updated checkboxes in Phase 2.md and overview.md.

Sessions 5-N - Continue Implementation (New Chat for each phase):

... repeat for each remaining phase ...

Final Session - Finalization (New Chat):

User: [Paste or invoke Step 4 prompt]
      [Attach: specs/task-api/ directory contents]

AI:   🧹 [VALIDATION]
      Verifying implementation...

AI:   🧹 [SPEC CLEANUP]
      Moving to specs/completed/task-api/

      Implementation complete!

---

Progress Tracking

The checkbox system enables seamless progress tracking across multiple sessions:

overview.md:

## Phases

- [x] Phase 1: Project Setup
- [x] Phase 2: Database Models
- [ ] Phase 3: API Endpoints <- Next phase to implement
- [ ] Phase 4: Authentication

Phase 3.md:

## Tasks

- [x] Create routes file
- [x] Implement GET /tasks
- [ ] Implement POST /tasks <- Current task
- [ ] Implement PUT /tasks/:id
- [ ] Implement DELETE /tasks/:id

---

Best Practices

1. Start fresh conversations - New chat for each step and each implementation phase
2. Always attach specs - The AI needs the spec files to understand the current state
3. Don't skip planning - The upfront investment prevents costly rework later
4. Confirm tech stack - Ensure AI gets explicit approval before architecture design
5. One phase at a time - Keeps conversations focused and manageable
6. Update checkboxes immediately - Maintains accurate progress state
7. Review phase output - Verify each phase before moving to the next
8. Keep spec files - The completed folder serves as project documentation

---

What to Attach at Each Step

Step	Required Attachments
Step 1 (Plan)	None (describe your feature/project)
Step 2 (Document)	specs/PLAN DRAFT.md or planning conversation
Step 3 (Implement)	specs/<feature>/overview.md + current Phase X.md
Step 4 (Finalize)	All files in specs/<feature>/ directory

---

File Structure After Complete Implementation

your-project/
├── specs/
│   ├── completed/
│   │   └── feature-name/
│   │       ├── overview.md      # Archived with completion summary
│   │       ├── Phase 1.md       # All checkboxes marked [x]
│   │       ├── Phase 2.md
│   │       └── ...
│   └── another-feature/         # In-progress feature
│       ├── overview.md
│       └── Phase 1.md
|
├── your project files...
└── README.md

---

Customization

Feel free to modify these prompts to fit your workflow:

• Add testing phases - Uncomment/add testing requirements in Step 2
• Adjust confidence threshold - Change the 90% threshold in Step 1
• Modify output structure - Customize the specs folder organization
• Add code review steps - Enhance Step 3 with additional review gates

---

Troubleshooting

AI jumps ahead to implementation during planning:

• The prompts explicitly forbid this, but if it happens, remind the AI: "Stay in planning mode. Do not write code yet."

AI doesn't know what to implement:

• Make sure you attached the overview.md and relevant Phase X.md files
• The AI needs these files to understand the current state and tasks

Lost progress between sessions:

• Check overview.md for phase status
• Review individual phase files for task completion status

AI not following spec exactly:

• Reference the specific phase document and ask it to re-read the requirements

Too many/few phases:

• Adjust during Step 2 (Documentation) - phases should represent logical groupings of work