VitePress

Appearance

Getting Started ​

Get up and running with M2JS in under 2 minutes! Choose your path:

  • AI Documentation - Transform code into LLM-friendly docs
  • Dead Code Detection - Find and safely remove unused code

Installation ​

bash
# Install globally
npm install -g @paulohenriquevn/m2js

# Verify installation
m2js --version
bash
# Install globally
yarn global add @paulohenriquevn/m2js

# Verify installation
m2js --version
bash
# Install from marketplace
code --install-extension paulohenriquevn.m2js-vscode

# Or search "M2JS" in VS Code Extensions

Choose Your Path ​

AI Documentation ​

bash
# Analyze a single TypeScript file
m2js UserService.ts

# Output will be saved as UserService.md

Dead Code Detection ​

bash
# Analyze your project for unused code
m2js src/ --detect-unused

# Generate configuration for better performance
m2js --init-config

# Get detailed help
m2js --help-dead-code

VS Code Extension ​

  1. Right-click any .ts, .tsx, .js, or .jsx file
  2. Select "Generate M2JS Documentation" or "Analyze Dead Code"
  3. View results in interactive webview
  4. Copy to clipboard for your AI assistant

Example Output ​

Let's see M2JS in action with a real TypeScript file:

Input File (UserService.ts) ​

typescript
export interface User {
id: number;
email: string;
name: string;
isActive: boolean;
}

/**
* Service for managing user operations
* Handles CRUD operations with validation
*/
export class UserService {
private users: User[] = [];

/**
* Create a new user with validation
* @param userData - User data to create
* @returns Promise resolving to created user
*/
async createUser(userData: Omit<User, 'id'>): Promise<User> {
const user: User = {
id: Date.now(),
...userData
};

this.users.push(user);
return user;
}

/**
* Find user by email address
* @param email - User email to search for
* @returns Promise resolving to user or null
*/
async findByEmail(email: string): Promise<User | null> {
return this.users.find(user => user.email === email) || null;
}
}

M2JS Output (UserService.md) ​

markdown
# UserService.ts

## Business Context
**Domain**: User Management (95% confidence)
**Framework**: TypeScript
**Patterns**: Service Layer, Repository Pattern
**Architecture**: Clean Architecture

## Architecture Insights
**Layer**: Service Layer
**Responsibility**: User CRUD operations with validation
**Dependencies**: None (self-contained)
**Data Flow**: In-memory storage with array-based operations

## Types

### User
```typescript
interface User {
id: number;
email: string;
name: string;
isActive: boolean;
}

Functions ​

createUser ​

typescript
async createUser(userData: Omit<User, 'id'>): Promise<User>

Purpose: Create a new user with validation Business Rules: Auto-generates ID using timestamp Usage Pattern: User registration workflow

findByEmail ​

typescript
async findByEmail(email: string): Promise<User | null>

Purpose: Find user by email address Usage Pattern: User lookup and authentication Returns: User object or null if not found


**Result**: Original file was 847 tokens, M2JS output is 267 tokens (**68% reduction**)

## Common Use Cases

### For AI Coding Assistants

```bash
# Generate optimized context for ChatGPT/Claude
m2js src/services/ --ai-enhanced --batch

# Copy results to AI assistant for better responses

For Code Maintenance ​

bash
# Quick cleanup before refactoring
m2js src/ --detect-unused

# Safe removal workflow
m2js src/ --detect-unused | grep "SAFE TO REMOVE" -A2

# CI/CD integration
m2js src/ --detect-unused --format json > dead-code-report.json

For Documentation ​

bash
# Generate project documentation
m2js src/ --batch --output docs/api/

# Include dependency graphs
m2js src/ --graph --mermaid

For Team Onboarding ​

bash
# Create comprehensive analysis for new developers
m2js src/ --ai-enhanced --architecture-insights --semantic-analysis

# Understand code health
m2js src/ --detect-unused

Configuration ​

Create a .m2jsrc file in your project root:

bash
# Generate configuration file
m2js --init-config

Example .m2jsrc:

json
{
  "deadCode": {
    "enableCache": true,
    "maxCacheSize": 1000,
    "chunkSize": 50,
    "showProgress": true,
    "format": "table"
  },
  "files": {
    "ignorePatterns": [
      "**/node_modules/**",
      "**/dist/**",
      "**/*.test.*",
      "**/*.spec.*"
    ],
    "maxFileSize": 10
  },
  "aiEnhanced": true,
  "businessContext": true,
  "outputDirectory": "docs/api"
}

Next Steps ​

AI Documentation

Dead Code Detection

Quick Examples ​

AI Documentation Workflow ​

bash
# 1. Transform code for AI
m2js UserService.ts

# 2. Copy output to ChatGPT/Claude
# 3. Get better AI responses with context

Dead Code Cleanup Workflow ​

bash
# 1. Analyze project
m2js src/ --detect-unused

# 2. Remove safe items immediately
# Follow "SAFE TO REMOVE" suggestions

# 3. Review medium-risk items
# Check "REVIEW BEFORE REMOVING" warnings

# 4. Keep high-risk items
# Document "HIGH RISK" decisions

Need Help? ​