A lightweight CLI tool for exporting and combining code files and documents into a single context file, perfect for working with Large Language Models (LLMs).

• Interactive File Selection: Navigate and select files using an intuitive terminal interface
• Document Support: Process Word documents (.docx), Excel spreadsheets (.xlsx), and PDF files (*.pdf)
• Smart File Filtering: Automatically identifies and filters non-processable files
• Text Extraction: Extracts text content from documents including tables and structured data
• Organized Output: Generates descriptive context files with clear file separators and metadata
• Project Statistics: Shows real-time statistics about selected files
• Native Experience: Uses platform-native keyboard shortcuts and navigation patterns

# Clone the repository
git clone https://github.com/yourusername/codestract.git
cd codestract

# Install dependencies
pip install -r requirements.txt

# Run locally
python -m codestract [directory_path]

Install the tool globally to use codestract from any directory:

# Clone and install globally
git clone https://github.com/yourusername/codestract.git
cd codestract
pip install -e .

# Now you can run from anywhere
codestract [directory_path]

To uninstall:

pip uninstall codestract

1. Navigate the file tree using arrow keys
2. Press Space to toggle file selection
3. Use Enter to expand/collapse directories
4. Press e to export selected files
5. Press q to quit

For non-interactive processing of all files in a directory:

xcodestract [directory_path]

This will:

• Recursively scan the directory for all processable files
• Automatically process text files and documents (*.docx, *.xlsx, *.pdf)
• Generate a context file without any user interaction
• Skip common build/cache directories (node_modules, pycache, .git, etc.)

Generated context files are saved in your current directory with:

• Intuitive naming: {directory_name}_context_{timestamp}.txt
• File metadata and statistics
• Clear file separators
• Project summary

• Python 3.8+
• Terminal with Unicode support
• For macOS: iTerm2, Kitty, or WezTerm recommended for best experience

• Text Files: All common text-based file formats (code, configuration, etc.)
• Documents:
  • Word documents (*.docx)
  • Excel spreadsheets (*.xlsx)
  • PDF files (*.pdf)

MIT License