read-file

Inspect datasets from files or URLs using DuckDB.

1|Updated Apr 3, 2026
One-click install
npx skills add https://github.com/nq-rdl/agent-extensions --skill read-file-nq-rdl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: read-file
Source: https://github.com/nq-rdl/agent-extensions/tree/main/skills/duckdb/read-file
Command: npx skills add https://github.com/nq-rdl/agent-extensions --skill read-file-nq-rdl

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Quickly read and profile datasets by treating a provided file path or URL as the target dataset, enabling schema discovery and sample extraction without writing code.

Core Features & Use Cases

  • Unified data reading for local and remote sources (CSV, JSON, Parquet, Avro, Excel, SQLite, and spatial formats)
  • Automatic format detection with a single DuckDB macro to read data
  • Quick insights: schema, sample rows, and row counts for data exploration

Quick Start

Use a dataset path or URL to inspect its structure and content, and obtain immediate insights.

Frequently Asked Questions about read-file

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

FAQPage Schema
How do I profile a CSV or Parquet file without writing code?

You can profile a CSV or Parquet file by providing its path or URL. The Skill uses a DuckDB macro to automatically detect the format, read the dataset, and return schema details, sample rows, and row counts.

Can I read remote data files from a URL for schema discovery?

Yes, you can read remote data files from a URL for schema discovery. The Skill supports remote sources by using optional DuckDB preloads for httpfs and cloud credentials to access and inspect the dataset.

What data formats does DuckDB support for quick data profiling?

For quick data profiling, DuckDB supports reading CSV, JSON, Parquet, Avro, Excel, SQLite, and spatial formats. A single macro auto-detects these formats and routes the read operation to extract schema and samples.

Do I need to install DuckDB to inspect an Excel or SQLite dataset?

Yes, you need DuckDB installed to inspect an Excel or SQLite dataset. The Skill requires DuckDB to execute its auto-detecting macro and extract schema and sample rows from your target files.

What is the best way to get row counts and sample rows from a JSON file?

The best way to get row counts and sample rows from a JSON file is to pass the file path to this Skill. It leverages DuckDB to automatically detect the JSON format and return immediate schema and sampling insights.

Are there limitations when reading spatial formats or Avro files with DuckDB?

Limitations when reading spatial formats or Avro files with DuckDB depend on environment setup. Users may need optional preloads for httpfs or cloud credentials to ensure remote sources are correctly accessed and profiled.