dataframely

Run `npx skills add Quantco/dataframely --all -g -y` in your terminal to install this skill globally for your AI agent.

It is a Python library that validates the schema and content of Polars data frames, making data pipelines more robust and type hints more informative.

Define a dy.Schema subclass with typed columns and rules, then call Schema.validate(df, cast=True) to check and cast your data frame.

Yes. Use Schema.filter to get valid rows plus a FailureInfo object describing which rows failed and which rules were violated.

Yes. The included SKILL.md teaches agents best practices for schemas, collections, validation, and testing when writing dataframely code.