Installation
Use Python 3.11 or newer and FFmpeg with libx264, the subtitles filter (libass), and loudnorm.
Install from PyPI into a virtual environment:
python3 -m venv .venv
source .venv/bin/activate
python -m pip install praisonai-clips
praisonai-clips doctor
praisonai-clips studio
Set the OPENAI_API_KEY environment variable before processing your own video. To use the Resolve MCP server, install praisonai-clips[live]; Kdenlive live editing also needs the separate bridge in the editor setup.
Install from source
git clone https://github.com/MervinPraison/praisonai-clips.git
cd praisonai-clips
python3 -m venv .venv
.venv/bin/pip install -e '.[dev]'
cp .env.example .env
Copy .env.example only on first setup; preserve an existing .env. Set OPENAI_API_KEY in that file. OpenAI usage is billed to your API account; source audio and sampled frames are sent to OpenAI.
FFmpeg on macOS
The Mac validation used Homebrew's ffmpeg-full because the installed standard build lacked subtitle rendering:
brew install ffmpeg-full
export PATH="$(brew --prefix ffmpeg-full)/bin:$PATH"
.venv/bin/praisonai-clips doctor
Check that subtitle_filter, loudnorm_filter and h264_encoder are true. doctor performs no paid API requests. On Debian/Ubuntu, install ffmpeg and python3-venv through apt.
If your virtual environment has no pip, uv pip install --python .venv/bin/python -e '.[dev]' is an alternative.