YOGIWAY

Code compression & prompt optimization for AI assistants - Save 40-50% on tokens when using Cursor, Windsurf, ChatGPT, or any AI coding assistant!

🚀 Features

⚡ Code Compression

• Compress Selection - Select code, compress, paste in AI chat
• Compress File - Compress entire file with one command
• Compress All Open Files - Compress all open files at once
• Compress Project - Compress entire project structure

🎯 Prompt Optimization

• Optimize Prompt - Reduce prompt tokens with intelligent optimization
• Count Tokens - Count tokens for any LLM provider (OpenAI, Anthropic, Google, etc.)
• Calculate Cost - Estimate costs for any model (GPT-5, GPT-4, Claude, etc.)
• Format for Provider - Format prompts for specific API requirements

⌨️ Keyboard Shortcuts

• Cmd+Shift+Y (Mac) / Ctrl+Shift+Y (Windows) - Compress selection
• Cmd+Alt+Y (Mac) / Ctrl+Alt+Y (Windows) - Compress file

📋 Auto-Copy to Clipboard

All compressed output is automatically copied to clipboard - just paste in your AI chat!

💰 Cost Savings

Example:

• Original: 50KB code = 12,500 tokens = $0.125
• Compressed: 25KB = 6,250 tokens = $0.0625
• Savings: 50% = $0.0625 per request

For 1,000 requests/day:

• Daily savings: $62.50
• Monthly savings: $1,875
• Yearly savings: $22,500!

📖 Usage

1. Compress Selection

1. Select code in editor
2. Press Cmd+Shift+Y (or right-click → "YOGIWAY: Compress Selection for AI")
3. Paste in Cursor/Windsurf/ChatGPT chat

2. Compress File

1. Open any file
2. Press Cmd+Alt+Y (or Command Palette → "YOGIWAY: Compress File for AI")
3. Paste in AI chat

3. Compress Project

1. Command Palette (Cmd+Shift+P)
2. Type "YOGIWAY: Compress Project Structure"
3. Wait for compression (shows progress)
4. Paste in AI chat

4. Compress All Open Files

1. Command Palette (Cmd+Shift+P)
2. Type "YOGIWAY: Compress All Open Files"
3. Paste in AI chat

5. Optimize Prompt

1. Select or open prompt text
2. Command Palette (Cmd+Shift+P)
3. Type "YOGIWAY: Optimize Prompt for AI"
4. Choose optimization mode (Standard, Aggressive, Semantic, Adaptive)
5. Paste optimized prompt in AI chat

6. Count Tokens & Calculate Cost

1. Select text
2. Command Palette → "YOGIWAY: Count Prompt Tokens" or "YOGIWAY: Calculate Prompt Cost"
3. Select provider and model
4. View token count and cost estimate

🎯 Why YOGIWAY?

• 40-50% token reduction on average
• Ultra-compact format designed for LLM prompts
• Handles deeply nested structures efficiently
• Preserves code structure and readability
• Works with any AI assistant (Cursor, Windsurf, ChatGPT, Claude, etc.)

🔧 Requirements

• VS Code 1.60.0 or higher

📝 License

MIT

🔗 Links

• npm Package: @yogesh0333/yogiway
• Prompt Library: @yogesh0333/yogiway-prompt
• Marketplace: VS Code Marketplace

Made with ❤️ for developers who want to save on AI costs