huggingface-datasets

Fetch metadata, paginate rows, search text, and retrieve parquet URLs from Hugging Face datasets.

Updated Jun 15, 2026
One-click install
npx skills add https://github.com/Andrew-Girgis/token-holdem --skill huggingface-datasets-andrew-girgis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: huggingface-datasets
Source: https://github.com/Andrew-Girgis/token-holdem/tree/main/.agents/skills/huggingface-datasets
Command: npx skills add https://github.com/Andrew-Girgis/token-holdem --skill huggingface-datasets-andrew-girgis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the need for a streamlined workflow for interacting with the Hugging Face Dataset Viewer API, enabling users to fetch metadata, paginate content, search text, apply filters, download URLs, and retrieve size or statistics for datasets.

Core Features & Use Cases

  • Metadata Fetching: Access subset/split metadata and paginate through rows of a dataset.
  • Text Search: Implement search functionality for text matching and filter rows based on predicates.
  • Download URLs: Retrieve links to parquet files and download data efficiently.
  • Use Case: Suppose you are working with a large dataset of news articles and need to extract information on articles that contain specific keywords, download the relevant parquet files, and perform further analysis.

Quick Start

Use the huggingface-datasets skill to search for news articles containing the keyword 'climate change' in the 'corpus' split of the 'news' dataset.

Frequently Asked Questions about huggingface-datasets

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

FAQPage Schema
How do I search and filter rows in a Hugging Face dataset?

You can search and filter Hugging Face datasets by using this Skill to query the Dataset Viewer API, applying text matching and specific predicates to extract matching rows efficiently.

Can I download parquet files directly from a Hugging Face dataset?

Yes, you can download parquet files directly by using this Skill to retrieve parquet file URLs from the Hugging Face Dataset Viewer API, enabling efficient data retrieval for local analysis.

Do I need an HF_TOKEN to fetch dataset metadata from the Hugging Face API?

Yes, you need an HF_TOKEN for authentication. This Skill requires internet access and a valid Hugging Face token to successfully fetch dataset metadata, paginate content, and retrieve statistics.

How do I paginate through large splits of a dataset on Hugging Face?

You can paginate through large dataset splits by leveraging this Skill's read-only access to the Hugging Face Dataset Viewer API, allowing you to iterate through rows and retrieve subset metadata.

What is the best way to get size and statistics for a Hugging Face dataset?

The best way to get size and statistics is using this Skill to query the Dataset Viewer API, retrieving dataset size details and statistics alongside metadata fetching and text search capabilities.

What are the limitations of read-only access to the Hugging Face Dataset Viewer API?

The limitation of read-only API access is that you cannot modify or upload datasets; this Skill strictly allows fetching metadata, paginating rows, searching text, filtering, and downloading parquet URLs.