Detects and corrects spelling errors in ETL metadata and snapshot files using codespell.
The Metadata Spell Checker skill is designed to maintain high-quality documentation within data pipelines by scanning .meta.yml and .dvc files for spelling errors. It integrates seamlessly with codespell, utilizing project-specific ignore lists and smart filtering to exclude archived datasets automatically. This skill is essential for data engineers and curators working on large-scale ETL projects like Our World in Data, providing an interactive workflow to review, ignore, or bulk-fix typos across thousands of files while respecting technical terminology.
Key Features
01Support for custom whitelists via .codespell-ignore.txt
02129 GitHub stars
03Automated typo detection in .meta.yml and .dvc files
04Interactive review and bulk-correction workflows
05Smart exclusion of archived steps and snapshots from checks
06Targeted scanning of specific ETL steps or full repository audits
Use Cases
01Batch correcting common misspellings across legacy data descriptors
02Maintaining professional documentation standards in large-scale data repositories
03Quality-checking metadata before publishing new datasets
