duckdb

Run SQL analytics directly on CSV, Parquet, and JSON files and Pandas or Polars DataFrames.

2|Updated Oct 24, 2025

Search Tags:#analytics #sql #csv #parquet #duckdb #dataframes

Ready to run

npx skills add https://github.com/silvainfm/claude-skills --skill duckdb

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: duckdb
Source: https://github.com/silvainfm/claude-skills/tree/main/duckdb
Command: npx skills add https://github.com/silvainfm/claude-skills --skill duckdb

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables fast, in-process SQL analytics directly on data files (CSV, Parquet, JSON) and Python DataFrames (pandas, Polars) without loading data into memory or running a separate database server.

Core Features & Use Cases

Direct file analytics: Run SQL queries on data files (CSV, Parquet, JSON) without importing.
DataFrame SQL: Query pandas or Polars DataFrames via SQL for seamless analytics workflows.
Advanced analytics: Use joins, aggregations, window functions, and CTEs on mixed data sources.
Ad-hoc exploration: Quickly explore datasets and generate insights without heavy setup.

Quick Start

Use this skill to run a simple SQL query on a CSV file, for example: SELECT category, AVG(price) AS avg_price FROM 'sales.csv' GROUP BY category

Frequently Asked Questions about duckdb

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I run SQL queries directly on CSV and Parquet files without loading them into a database?▼

SQL queries on CSV, Parquet, and JSON files execute in-process without importing data into memory or running a separate server. DuckDB provides a file-first query engine that reads data directly from files, applies SQL operations like aggregations and joins, and returns results immediately for ad-hoc exploration.

Can I query pandas and Polars DataFrames using SQL syntax?▼

Yes, pandas and Polars DataFrames can be queried directly with SQL. DuckDB accepts DataFrames as data sources, enabling you to write SQL queries—including joins, window functions, and CTEs—against DataFrame data without converting to another format.

What's the best way to perform analytics on mixed data sources like CSV files and DataFrames together?▼

Cross-source joins combine CSV, Parquet, JSON files and pandas or Polars DataFrames in a single SQL query. DuckDB treats all sources uniformly, allowing you to join files to DataFrames, aggregate across sources, and retrieve results without manual data consolidation.

Do I need to set up a database server to run SQL analytics on local data files?▼

No database server is required. DuckDB runs as an in-process SQL engine on your local machine, querying files and DataFrames directly without external infrastructure. Setup is minimal—queries execute immediately against your data sources.

Can I use window functions and CTEs for complex analytical queries on files?▼

Yes, DuckDB supports advanced SQL features including window functions, common table expressions (CTEs), and subqueries. These enable complex aggregations, ranking, and multi-step analytics on CSV, Parquet, JSON files and DataFrames within a single query.

What file formats does SQL-based analytics work with?▼

CSV, Parquet, and JSON files are supported for direct SQL queries. DuckDB also works with pandas and Polars DataFrames, enabling analytics across multiple file formats and in-memory data structures in a single workflow.