Transforms and analyzes large datasets using DuckDB SQL directly within the Claude Code environment.
The Data Wrangler skill empowers Claude to perform complex data engineering and analytics tasks by leveraging the high-performance DuckDB SQL engine. It allows users to query, join, and transform data across diverse formats including CSV, Parquet, JSON, and Excel, as well as external databases like PostgreSQL and MySQL. By offloading data processing to a dedicated engine, it provides a robust solution for cleaning datasets, creating partitioned files, and generating samples, making it significantly more efficient and accurate than in-context reasoning for high-volume or complex data tasks.
Key Features
01Advanced transformations including PIVOT, UNPIVOT, and window functions
02Multi-source joins across local files and remote databases
031 GitHub stars
04High-performance SQL analytics via integrated DuckDB engine
05Flexible export capabilities to Parquet, CSV, and JSON with partitioning
06Automated data profiling, schema detection, and statistics
Use Cases
01Cleaning and filtering massive CSV or Parquet files for analysis
02Converting and partitioning raw data into optimized formats like Parquet
03Joining local datasets with remote PostgreSQL or MySQL database tables
