quant-ranger

Automated repository maintenance across GitHub organizations. quant-ranger discovers applicable work, runs typed update tasks in fresh checkouts, and can publish the results as pull requests.

Getting started

Run the CLI directly from conda-forge with Pixi:

pixi exec quant-ranger --help

For repeated use, install it globally:

pixi global install quant-ranger

Run an updater in dry-run mode with an existing GitHub CLI login:

quant-ranger update --gh --repository octo-org/octo-repo pixi-update

Add --publish-changes before the updater name to create or update a pull request.

See the documentation for updaters, authentication, automation, extension APIs, and the CLI reference.

Installing from PyPI

Installing quant-ranger as a conda package is strongly recommended because quant-ranger drives several external command-line tools that only conda can declare as dependencies. The wheel on PyPI contains the same Python code, but if you install it with pip, you must make the tools listed under [package.run-dependencies] in pixi.toml available on your PATH.

Missing tools surface as runtime failures in the updaters that need them rather than as install-time errors.

The conda distribution is fully self-contained.

Development

Everything runs from the source checkout through Pixi. The first pixi run creates the environment:

git clone https://github.com/quantco/quant-ranger
cd quant-ranger
pixi run pre-commit-install
pixi run test
pixi run lint

Run the CLI from the checkout with pixi run quant-ranger .... Documentation tasks are described in docs/README.md.