OmegaFlow turns scripted terminal workflows into rebuildable videos for docs, tutorials, release notes, and technical demos.
Instead of recording a one-off screen capture, you write a versioned Markdown recording script. The script contains prose for readers, YAML directives for OmegaFlow, terminal commands to run, expected output checks, optional generated voiceover, and publish targets. OmegaFlow records the terminal session, lines it up with the script and narration, and writes website-ready assets.
Use it when a demo should live with the code it explains:
- rebuild videos when commands, docs, or setup steps change
- keep terminal demos reviewable in source control
- generate optional voiceover from the same recording script
- publish videos as plain HTML or embedded Docusaurus pages
Watch the homepage demo · View its recording source
Recording is supported on Linux and macOS. On Windows, use WSL for the Linux recording workflow.
Install OmegaFlow in your project's Python environment:
python -m pip install omegaflowOmegaFlow requires Python 3.11+ and Bash. Supported Linux and macOS wheels include asciinema 3.x.
Create the initial recording workspace:
omegaflow bootstrap=projectThis creates the project configuration and a small recording script at
recordings/test-video/index.md. Edit that Markdown file to define the prose,
commands, checks, and optional narration for your video.
Build the generated test video:
omegaflow recording=test-videoWatch the generated terminal video and optional narration locally:
omegaflow recording=test-video action=watchContinue with Build your first video or browse the full documentation.
See the maintainer guide for development setup, local website instructions, and validation.
OmegaFlow is available under the Mozilla Public License 2.0. Source for released versions is available from the OmegaFlow repository. See NOTICE.md for relicensing and third-party notices.