TidyPanel - Universal Messy Panel Data Cleaner
A robust toolkit designed to standardize and clean complex tabular data from commercial enterprise systems, healthcare records, logistics software, and HR databases. Now natively supports Excel files, flat-file databases (CSV/TSV/RDB), gzip-compressed CSV/TSV/TXT/RDB files, and raw data.frame inputs. Features include intelligent regex parsing for domain-specific noise (currencies, percentages), horizontal side-by-side panel splitting, automated metadata key-value extraction from decoy rows, gap-based block clustering, and messy table resolution. Methods draw on tidy data principles described in Wickham (2014) <doi:10.18637/jss.v059.i10> and the 'readxl' parsing infrastructure described in Wickham & Bryan (2023) <https://readxl.tidyverse.org>. A local JSON command-line interface supports reproducible table-cleaning jobs and bounded agent handoffs.
Last updated
agent-toolsai-agentscommand-linecsvdata-cleaningdata-wranglingexceljsonllm-toolslocal-firsttabular-datatoken-saving
4.70 score 1 stars 236 downloads