Create viral-style "brainrot" videos from GitHub repository README files. Generates a summarized script, text-to-speech narration, and composites it over gameplay footage with captions.
- Python 3.12 (required - Python 3.13 is not supported due to dependency compatibility)
- uv (recommended package manager)
- FFmpeg (for video processing)
- Gemini API key
curl -LsSf https://astral.sh/uv/install.sh | shuv venv --python 3.12 .venvsource .venv/bin/activateuv pip install -r requirements.txtCopy the example env file and add your Gemini API key:
cp .env.example .envEdit .env and add your key:
GEMINI_API_KEY=your_gemini_api_key_here
Create a backgrounds/ folder and add gameplay videos (Subway Surfers, Minecraft parkour, satisfying clips, etc.):
mkdir -p backgrounds
# Add your .mp4 files to this folderpython main.py https://github.com/facebook/react
python main.py anthropics/claude-code --voice Puckpython main.py --readme path/to/README.md
python main.py -r oai_readme.md --output openai_brainrot.mp4python main.py --list-voices| Option | Description |
|---|---|
repo |
GitHub repository URL or owner/repo format |
--readme, -r |
Path to a local README file |
--voice |
TTS voice to use (default: Puck) |
--output, -o |
Output filename |
--backgrounds |
Directory with background videos (default: backgrounds/) |
--skip-summary |
Use raw README without AI summarization |
--list-voices |
Show all available voices |
If you see errors about ctranslate2, llvmlite, or missing wheels for cp313, you need to use Python 3.12:
# Check your Python version
python --version
# If using Python 3.13, create a new venv with 3.12
uv venv --python 3.12 .venv
source .venv/bin/activate
uv pip install -r requirements.txtInstall FFmpeg:
# macOS
brew install ffmpeg
# Ubuntu/Debian
sudo apt install ffmpegVideos are saved to the output/ directory by default.