# Contributing to Modelship

Thanks for your interest in contributing to Modelship! This document covers the basics for getting started.

## License

Modelship is licensed under [Apache 2.0](LICENSE), and **will always remain available under Apache 2.0** — that promise isn't contingent on the CLA below.

Per Apache 2.0 §5, any contribution you intentionally submit for inclusion is licensed to the project under Apache 2.0 by default. Before your first PR can be merged, you'll also need to sign the [Contributor License Agreement](CLA.md) (individual or corporate) via the CLA Assistant bot, which will comment on your PR with a link. You keep copyright in your own contributions — the CLA grants the project the right to distribute and, if ever needed, relicense them, it does not assign them away.

## Development Setup

The recommended way to develop is via the VS Code Dev Container. See [docs/development.md](docs/development.md) for full instructions.

**Quick version:**

1. Install Docker + NVIDIA Container Toolkit
2. Set the `HF_TOKEN` environment variable
3. Open the project in VS Code and select **Dev Containers: Reopen in Container**
4. Inside the container:

```bash
uv sync --extra dev
ray start --head --port=6379 --dashboard-host=0.0.0.0 --num-cpus=2 --num-gpus=1
uv run mship_deploy.py
```

## Code Style

The project uses [Ruff](https://docs.astral.sh/ruff/) for linting and formatting, and [Pyright](https://github.com/microsoft/pyright) for type checking. Both run in CI on every pull request.

```bash
ruff check .          # lint
ruff format .         # format
pyright               # type check
```

Configuration is in `pyproject.toml`. Key settings: Python 3.12, 120 character line length.

## Submitting Changes

1. Fork the repository and create a branch from `main`
2. Make your changes
3. Ensure `ruff check`, `ruff format --check`, and `pyright` pass
4. Open a pull request against `main` with a clear description of what changed and why

## Writing Plugins

If you're contributing a TTS backend, see [docs/plugins.md](docs/plugins.md) for the plugin architecture and development guide. Plugins live in `plugins/` as isolated uv workspace packages.

## Reporting Issues

Use [GitHub Issues](https://github.com/modelship-ai/modelship/issues). For bugs, include:

- GPU model and VRAM
- Your `models.yaml` configuration
- Docker and NVIDIA driver versions
- Relevant logs from the container

## Security

For security vulnerabilities, see [SECURITY.md](SECURITY.md) for responsible disclosure instructions.