🎬 Clawcut

Clawcut is a modular, agentic video editing engine designed for the Clawhub AI ecosystem. It provides a "Sat-Set" (fast and efficient) alternative to heavy editing frameworks by leveraging a declarative Python wrapper for FFmpeg.

---

🚀 Why Clawcut?

• Modular Architecture: Easy to extend filters, presets, and engine logic.
• Zero-Config: No complex system dependencies. Just FFmpeg and Python.
• Agentic-First: Designed to be controlled by AI agents via JSON presets and CLI.
• Fast ("Sat-Set"): Native FFmpeg execution for high-speed rendering.

📂 Project Structure

• main.py: CLI entry point.
• clawcut/: Core package containing engine, filters, and utilities.
• presets/: Folder for layout definitions (JSON).
• assets/branding/: Place your watermark.png and poppins-bold.ttf here.
• outputs/: Default directory for rendered videos.

🛠 Installation

1. Requirements:

   • FFmpeg
   • yt-dlp
   • python-dotenv (optional)

2. Setup:

   git clone https://github.com/pendig/clawcut.git
   cd clawcut
   pip install -r requirements.txt

🏁 Quick Start

Render a branded Podcast Clip:

python3 main.py --url "YOUTUBE_URL" --preset podcast-clip --title "Modular Power" --watermark

---

📄 License

MIT License. Created by Pena Digital.