CodePacker AI

Codebase to TXT & Clipboard for AI Models

Developed by KiranLabs

Supercharge your AI development workflow. Export entire codebases or selected folders into a structured, token-efficient .txt file or instant clipboard copy.

🌟 Why CodePacker AI?

When using AI models like ChatGPT, Claude 3.5, Gemini 1.5 Pro, or DeepSeek, the quality of their answers depends directly on giving them clean, complete codebase context. Manually copying files is slow, tedious, and prone to missing critical imports.

CodePacker AI solves this instantly:

⚡ One-Click Export: Right-click any directory, file, or export your whole workspace.
📋 Direct to Clipboard: Instantly copies with statistics, ready to Ctrl + V into your AI chat.
🌳 Automatic Project Tree: Generates a clean ASCII file tree at the top of the context so the AI understands your directory structure and file hierarchy.
🛡️ Intelligent Filtering: Automatically respects .gitignore, skips node_modules, .git, build output, lock files, and binary files (images, audio, PDFs, archives).
📊 Token & Line Stats: Gives you accurate file, line, and estimated token counts before sending.

🚀 Quick Usage

1. File Explorer (Right-Click)

Right-click on any folder or file in VS Code's sidebar and choose:

CodePacker AI: Export to .txt File and Copy to Clipboard (Recommended)
CodePacker AI: Export Codebase to .txt File
CodePacker AI: Copy Codebase to Clipboard

2. Command Palette

Press Ctrl + Shift + P (or Cmd + Shift + P on Mac) and type:

CodePacker AI

📄 Output Preview

Here is an example of what CodePacker AI generates (with folder locations and code line numbers):

================================================================================
CODEBASE EXPORT FOR AI
Project: my-project
Generated: 2026-09-27 23:38:00
Total Files: 3
Total Lines: 120
Total Size: 4.2 KB
Estimated Tokens: ~1,050 tokens
================================================================================

================================================================================
PROJECT DIRECTORY & FOLDER STRUCTURE:
================================================================================
└── 📁 src/
    ├── 📁 components/
    │   └── 📄 Button.jsx
    ├── 📄 index.js
    └── 📄 utils.js

================================================================================
FILE CONTENTS (WITH LINE NUMBERS & LOCATIONS):
================================================================================

================================================================================
FILE: src/components/Button.jsx
LOCATION: src/components/
FILENAME: Button.jsx
LINES: 24 | SIZE: 0.85 KB
--------------------------------------------------------------------------------
   1 | import React from 'react';
   2 | 
   3 | export const Button = ({ label, onClick }) => {
   4 |   return (
   5 |     <button className="btn-primary" onClick={onClick}>
   6 |       {label}
   7 |     </button>
   8 |   );
   9 | };

⚙️ Settings & Customization

Configure preferences under VS Code Settings (Ctrl + , -> search CodePacker AI):

Setting	Default	Description
`aiCodePacker.outputFileName`	`codebase_ai_context.txt`	Base name for exported file
`aiCodePacker.appendTimestamp`	`true`	Creates unique date/time stamped file on each run
`aiCodePacker.includeLineNumbers`	`true`	Aligned line numbers while strictly preserving code indentation
`aiCodePacker.outputFormat`	`txt`	Format: `'txt'` (clean ASCII dividers) or `'markdown'` (fenced code blocks)
`aiCodePacker.includeDirectoryTree`	`true`	Include visual directory tree with folders and files
`aiCodePacker.respectGitignore`	`true`	Automatically parse `.gitignore` rules
`aiCodePacker.maxFileSizeKB`	`1024`	Skip files exceeding this limit (in KB)
`aiCodePacker.additionalIgnorePatterns`	Lock files & minified files	Extra glob patterns to exclude

📦 About KiranLabs

KiranLabs builds developer tooling to make programming with AI fast, seamless, and reliable.

More extensions from KiranLabs coming soon!

📄 License

MIT License. Free to use, modify, and distribute.

CodePacker AI: Codebase to TXT for AI

KiranLabs