Skip to content

A lightweight Python tool for extracting codebase context for LLM conversations - generate directory trees and file contents for better AI interactions

License

Notifications You must be signed in to change notification settings

ergut/contextor

Repository files navigation

Contextor 🚀

Here's a secret about AI coding assistants: they're only as good as the context you give them! Forget chasing perfect prompts or waiting for the next big model - what truly transforms an AI assistant into a reliable coding partner is crystal-clear context about your project.

Ever needed to explain your codebase to ChatGPT or Claude? Contextor creates a perfect snapshot of your project in seconds:

# That's it! Just run:
contextor --directory ./my_project

What You Get ✨

my_project/
├── src/
│   └── main.py     # LLMs can request this file if needed!
└── config/
    └── settings.yaml

# Key files are included below the tree...

Just paste this into your AI chat and start coding! The AI can see your project structure and request any file it needs.

Quick Start 🏃‍♂️

# Install
pip install contextor

# Run (will include all files)
contextor --directory ./my_project

# Or specify key files only
contextor --files main.py config.yaml

Why Contextor? 🎯

  • Simple: One command to create perfect context for AI conversations
  • Smart: Respects .gitignore, handles large files, includes safety checks
  • Flexible: Include specific files or let the AI see everything
  • Safe: Warns you about size and skips files >10MB

Features in Detail 🛠️

  • 📁 Complete project tree generation
  • 📄 Automatic or selective file inclusion
  • 🔒 .gitignore pattern support
  • ⚡ Large file protection
  • 🎮 Custom file exclusions
  • 📊 Size warnings and confirmations

Advanced Usage 🔧

Need more control? We've got you covered:

# Include files listed in a text file
contextor --files-list important_files.txt

# Custom exclude patterns
contextor --exclude-file exclude_patterns.txt

# Ignore .gitignore
contextor --no-gitignore

# Include essential context and supplementary info
contextor --prefix-file project_overview.txt --appendix-file api_docs.txt

# Add schemas and deployment guides
contextor --prefix-file schemas.txt --appendix-file deployment.txt

Command Line Options 🎛️

Option Description
--directory Project directory (default: current)
--files Specific files to include
--files-list File containing list of files
--smart-select Automatically select important files like entry points, configs, and docs
--prefix-file Essential context to add at start (schemas, overview)
--appendix-file Supplementary info to add at end (docs, guides)
--output Output filename (default: project_context.txt)
--estimate-tokens Calculate and show estimated token count in the output file
--no-gitignore Disable .gitignore patterns
--exclude-file Additional exclude patterns file

Examples 📚

Include specific files (files-list.txt):

src/main.py
config/settings.yaml
README.md

Exclude patterns (exclude-patterns.txt):

*.pyc
__pycache__/
.env
*.log

Safety First 🛡️

Contextor looks out for you:

  • Calculates total file size
  • Shows warning for large directories
  • Asks for confirmation
  • Skips files >10MB
  • Respects .gitignore by default

Installation Options 📦

# From PyPI (recommended)
pip install contextor

# From source
git clone https://github.com/ergut/contextor
pip install -r requirements.txt

Contributing 🤝

We love contributions! Check out README.test.md for:

  • Running tests
  • Test coverage details
  • Adding new features
  • Contributing guidelines

License 📜

MIT License - See LICENSE file

Support 💬

Author ✍️

Salih Ergüt

Version 📋

Current version: 1.0.3

See CHANGELOG.md for version history and updates.

About

A lightweight Python tool for extracting codebase context for LLM conversations - generate directory trees and file contents for better AI interactions

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages