Files
live/docs/contributing.md
zeekay d7f78520cc Rename package from daydream-scope to hanzo-live
This commit renames the Python package to better align with Hanzo AI branding.

Changes:
- Package name: daydream-scope → hanzo-live
- Command: daydream-scope → hanzo-live
- Models directory: ~/.daydream-scope/models → ~/.hanzo-live/models
- Environment variable: DAYDREAM_SCOPE_MODELS_DIR → HANZO_LIVE_MODELS_DIR
- Updated all documentation and references
- Updated authors to Hanzo AI
- Added "hanzo" and "mlx" keywords

Installation:
  pip install hanzo-live

Usage:
  hanzo-live            # Start the server
  hanzo-live --cpu      # CPU-only mode
  hanzo-live --version  # Show version
2025-10-20 22:13:04 -07:00

1.8 KiB

Contributing

Thank you for your interest in contributing!

We use the Developer Certificate of Origin (DCO) to ensure that all contributions can be legally included and relicensed in future releases.

To certify your contribution, please sign off each commit with:

git commit -s

This adds a line like this to your commit message:

Signed-off-by: Your Name <[email protected]>

By signing off, you confirm that you have the right to submit this code and that it may be distributed under the same license as this project.

Setup

Install uv which is needed to run the server, Node.js which is needed to build frontend and ffmpeg which is needed for certain media workflows during development.

Frontend

First, navigate to the frontend directory.

Install dependencies:

npm install

Run the development server (hot reloading is on by default):

npm run dev

Server

Install all (including development) dependencies:

uv sync --group dev

Install pre-commit hooks:

uv run pre-commit install

Run the server with hot reloading enabled:

uv run hanzo-live --reload

Testing Pipelines

By default, the server does not load any pipelines on startup, but you can set the PIPELINE environment variable to automatically load a specific pipeline on startup which can be useful for testing.

This would load the longlive pipeline on startup:

PIPELINE="longlive" uv run hanzo-live

You can also test the longlive pipeline on its own:

uv run -m pipelines.longlive.test

This test outputs a video file.

Release Process

TBD