LLM Context Builder - Pip Package Structure

Package Structure

llm-context-builder/
├── setup.py
├── pyproject.toml
├── README.md
├── llm_context_builder/
│   ├── __init__.py
│   ├── main.py
│   ├── project_detector.py
│   └── exporters/
│       ├── __init__.py
│       └── base_exporter.py
└── requirements.txt

Installation & Usage

Installation

pip install llm-context-builder

Usage

# Export current directory
export_project

# Export specific directory
export_project /path/to/project

# Custom output location
export_project -o /custom/path/output.txt

# Quick web project export (auto-detects React/Vue/etc.)
export_project --preset web

# Quick Python project export
export_project --preset python

# Show token count estimate
export_project --count-tokens

# List available presets
export_project --list-presets

Features

🎯 Smart Project Detection

• Automatically detects project type (Python, Node.js, React, Vue, etc.)
• Applies appropriate exclusions for each project type
• Suggests optimal presets

🧠 LLM-Optimized Output

• Clean, readable format perfect for LLM context
• Token counting to estimate context usage
• Optimized file headers for better parsing
• Smart content truncation for large files

⚡ Zero Configuration

• Works out of the box in any directory
• Sensible defaults for common project types
• Creates project_export/ in current directory

🎛 Highly Configurable

• Custom exclusion patterns
• Project-specific presets
• Flexible output options
• Override any default behavior

Example Outputs

Command:

cd ~/my-react-app
export_project

Result:

✅ Detected: React project
📁 Output: ./project_export/react_app_20250602_143052.txt
📊 Files processed: 45
🔤 Estimated tokens: ~12,500
⏱ Completed in 0.3s

Command:

export_project --preset python --count-tokens

Result:

🐍 Using Python preset
📁 Output: ./project_export/python_project_20250602_143105.txt
📊 Files: 23 processed, 156 skipped
🔤 Tokens: 8,342 (fits in GPT-4 context)
📋 Included: .py, .md, .txt, .yml, .toml
🚫 Excluded: __pycache__, .git, venv, *.pyc

Project Presets

Available Presets:

• web - React, Vue, Angular, general web projects
• python - Python packages, Django, Flask projects
• node - Node.js, npm packages
• mobile - React Native, Flutter
• docs - Documentation projects
• minimal - Only essential text files
• full - Include everything (use with caution)

Each preset automatically configures:

• File type inclusions/exclusions
• Folder exclusions
• Size limits
• Output formatting