lintquarto

Project info:
Installation:
Metrics:
Build & quality status:
Supported platforms:

Package for running linters, formatters, static type checkers and code analysis tools on python code in quarto (.qmd) files.

By default, python code validation tools can't check embedded python code in Quarto files. This package fills that gap, enabling analysts and researchers to run python quality checks within Quarto documents.

Currently supported:

• Linters: pylint, flake8, pydoclint, pyflakes, ruff, vulture, and pycodestyle.
• Static type checkers: basedpyright, mypy, pyright, pyrefly, and pytype.
• Code analysis tools: radon.
• Formatters: ruff.

This package is featured on awesome-quarto.

This package has been reviewed by PyOpenSci.

Installation

You can install lintquarto with pip (from PyPI or GitHub) or conda (from conda-forge).

Install with pip

pip install lintquarto

To include your selection of linters, install them as needed.

For a one-step installation that includes lintquarto and all supported tools, use:

pip install lintquarto[all]

Install with uv

uv add lintquarto

Or to include all supported tools:

uv add "lintquarto[all]"

Install with conda

conda install conda-forge::lintquarto

With conda, only the main lintquarto tool is installed. If you want to use any linters or type checkers, you must install them separately (either with conda or pip, depending on availability).

Development version

To install the latest development version of lintquarto directly from this repository:

pip install git+https://github.com/lintquarto/lintquarto

If you also want all supported code quality tools in one step, install from a local clone in editable mode with the all extra:

git clone https://github.com/lintquarto/lintquarto.git
cd lintquarto
pip install -e ".[all]"

Getting started using lintquarto

Usage

Usage:

lintquarto [-h] [-l LINTER [LINTER ...]] [-f FORMATTER [FORMATTER ...]] [-p PATHS [PATHS ...]] [-e [[exclude_paths] ...]] [-n] [-v] [-k] [-c COMMAND] {list} ...

Lint Python code in Quarto (.qmd) files. Configuration can also be provided in pyproject.toml under [tool.lintquarto]. CLI arguments override configuration file.

Options:

• -h, --help - show this help message and exit
• -l, --linters LINTER [LINTER ...] - Linters to run. Valid options: ['basedpyright', 'flake8', 'mypy', 'pycodestyle', 'pydoclint', 'pyflakes', 'pylint', 'pyright', 'pyrefly', 'pytype', 'radon-cc', 'radon-mi', 'radon-raw', 'radon-hal', 'ruff', 'vulture']
• -f, --formatters FORMATTER [FORMATTER ...] - Formatter to run. Valid options: ['ruff-format', 'ruff-check-fix'].
• -p, --paths PATHS [PATHS ...] - Quarto files and/or directories to run tools on.
• -e, --exclude [[exclude_paths] ...] - Files and/or directories to exclude from running tools on.
• -n, --lint-non-exec - Also lint non-executable Python code chunks
• -v, --verbose - Verbose output.
• -k, --keep-temp - Keep temporary .py files after linting.
• -c, --custom-commands COMMAND - Custom command to run against the generated .py file. Repeat for multiple commands. Example: --custom- commands "mytool"

Commands:

• list - List supported linters and whether they are available.

Passing extra arguments directly to linters is not supported. Only .qmd files are processed.

Configuration file

As an alternative to passing flags on every run, you can declare your settings once in a [tool.lintquarto] section in your pyproject.toml. The arguments are equivalent to those used on the command line. For example:

[tool.lintquarto]
linters = [
  "ruff",
  "pycodestyle",
]
paths = [
  "examples/",
  "dashboard/index.qmd",
]
lint-non-exec = false

With this in place, you can run lintquarto with no arguments.

Note: CLI flags will always take priority over pyproject.toml. If you supply -l or -p on the command line, those values are used and the corresponding config file values are ignored. exclude and custom-commands are additive - values from both sources are merged together.

Examples

The linter used is interchangeable in these examples.

Lint all .qmd files in the current directory (using pylint):

lintquarto -l pylint -p .

Lint several specific files (using pylint and flake8):

lintquarto -l pylint flake8 -p file1.qmd file2.qmd

Keep temporary .py files after linting (with pylint)

lintquarto -l pylint -p . -k

Lint all files in current directory (using ruff):

• Excluding folders examples/ and ignore/, or-
• Excluding a specific file analysis/test.qmd.

lintquarto -l ruff -p . -e examples ignore

lintquarto -l ruff -p . -e analysis/test.qmd

Find out more

Visit our website to find out more and see examples from running with each code validation tool.

Community

Thanks goes to these wonderful people (emoji key):

Amy Heather 🎨 📖 🤔 🚇 🚧 ⚠️ 🐛 💻	Isabel Zimmerman 🤔 📖	Jacob Cumming 🐛 📖 🤔 ⚠️ 💻	Jonny Saunders 🐛 💻 🤔
Sammi Rosser 💻 📖 ⚠️	Anna Laws 📖	Nissim Lebovits 🤔	Tom Monks 📖

This project follows the all-contributors specification. Contributions of any kind welcome!

Curious about contributing? Check out the contributing guidelines to learn how you can help. Every bit of help counts, and your contribution - no matter how minor - is highly valued.

How to cite lintquarto

Please cite the repository on GitHub, PyPI, conda and/or Zenodo:

Heather, A. (2026). lintquarto (v0.14.0). https://github.com/lintquarto/lintquarto.

Heather, A. (2026). lintquarto (v0.14.0). https://pypi.org/project/lintquarto/.

Heather, A. (2026). lintquarto (v0.14.0). https://anaconda.org/conda-forge/lintquarto.

Heather, A. (2026). lintquarto (v0.14.0). https://doi.org/10.5281/zenodo.15731161.

Citation instructions are also provided in CITATION.cff.

Acknowledgements

This project was written and maintained by hand, with occasional use of Perplexity AI during development (specific models and versions varied over time).

AI assistance was used for small, targeted tasks (e.g. help when troubleshooting, identifying issues, refining docstrings, improving code structure), rather than to generate complete functions or substantial content.

All code and design decisions were reviewed and finalised by a human. For transparency, the use of AI is acknowledged, but the project should not be considered AI‑generated.