What problem does it solve? Python files generated or edited during a data-science workflow often ship with inconsistent formatting, lint violations, missing docstrings, and leftover template or workflow commentary. This Skill standardizes style enforcement with ruff and numpydoc so committed files read like a colleague wrote them. ## Core Features & Use Cases - Ruff lint and format pipeline: Runs ruff format, ruff check --fix, and ruff check in order on every Python file touched in a turn, with a one-fix-per-file rule that caps retries and surfaces persistent diagnostics. - Project ruff.toml setup: Drops a bundled ruff.toml template (line length 88, numpy pydocstyle convention, per-file ignores for experiments/, audit/, and data/eda.py) into fresh projects, always read verbatim from the template rather than authored from memory. - Comment contextualization: Rewrites leftover template and workflow prose in touched files into problem-specific documentation grounded in the dataset and experiment context. - Use Case: After generating an experiment script for a breast-cancer classifier, invoke this Skill to format the file, fix import ordering, add numpydoc docstrings to public functions, and strip scaffolding comments before committing. ## Quick Start Run ruff format and check on the Python files you just edited and rewrite their comments to describe the data-science problem.
