Inspect, profile, and query local or S3 data files with DuckDB.
npx skills add https://github.com/ejoliet/claude-skills --skill data-tools
Please help me install this Agent Skill. Skill: data-tools Source: https://github.com/ejoliet/claude-skills/tree/main/data-tools Command: npx skills add https://github.com/ejoliet/claude-skills --skill data-tools
This skill makes it fast and easy to understand data files before building pipelines by enabling quick schema discovery, sampling, and lightweight querying.
Ask me to inspect a file path or S3 URI and I will reveal the schema, sample rows, and a quick data profile.
High-intent search queries and answers about installing and using this skill.
You can inspect and profile S3 data files by providing an S3 URI, which triggers in-process DuckDB to perform schema sniffing, row sampling, and quick profiling on Parquet, CSV, or JSON formats.
Yes, you can run ad-hoc SQL queries on local Parquet and CSV files using DuckDB core patterns, enabling lightweight data exploration and validation before committing to a full pipeline.
DuckDB supports schema sniffing for FITS files with optional support from astropy Table, allowing you to sample rows and profile astronomical data formats alongside standard CSV and JSON files.
The best way to validate datasets before building pipelines is using lightweight CLI patterns to perform schema discovery and ad-hoc SQL queries, ensuring data structure and content meet requirements early.
Yes, pandas and awswrangler provide optional support for AWS and format-specific workflows, complementing DuckDB core patterns to handle complex data inspection and querying tasks across S3.
In-process DuckDB is designed for quick schema discovery and lightweight querying rather than heavy pipeline execution, so users should anticipate performance constraints when profiling extremely large-scale datasets.
ryankoleanRun in-process analytical SQL queries on CSV, Parquet, and JSON files.
nahisahoSearch HCA Data Portal projects and export CELLxGENE Census subsets for Scanpy analysis.
cas-bigdatalabCompare data table field names and counts against a predefined schema.
crossxwillQuery Parquet data with DuckDB and convert results to pandas DataFrames.
