hugging-face-datasets

Manage Hugging Face datasets with SQL-based querying and transformation.

1|Updated Feb 24, 2026
One-click install
npx skills add https://github.com/FacuM/yolo-agent --skill hugging-face-datasets-facum
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hugging-face-datasets
Source: https://github.com/FacuM/yolo-agent/tree/main/.claude/skills/hugging-face-datasets
Command: npx skills add https://github.com/FacuM/yolo-agent --skill hugging-face-datasets-facum

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires duckdb, huggingface_hub, datasets, pandas, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines the creation, management, and querying of datasets on the Hugging Face Hub, enabling efficient data preparation and manipulation for AI models.

Core Features & Use Cases

  • Dataset Lifecycle Management: Initialize new repositories, configure system prompts, and stream row updates.
  • SQL-Based Querying: Query, transform, and analyze any Hugging Face dataset using SQL via DuckDB.
  • Multi-Format Support: Handles diverse dataset types including chat, classification, QA, text completion, and tabular data.
  • Use Case: You can query a large dataset using SQL to filter for specific examples, transform them into a new format, and then push the results to a new dataset repository on Hugging Face Hub.

Quick Start

Use the hugging-face-datasets skill to query the 'cais/mmlu' dataset for nutrition-related subjects.

Frequently Asked Questions about hugging-face-datasets

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

FAQPage Schema
How do I query a Hugging Face dataset using SQL?

You can query Hugging Face datasets using SQL via DuckDB, enabling direct filtering, transformation, and analysis of large datasets. This integrates with the HF MCP server to streamline data extraction without requiring full local downloads.

Can I push transformed data to a new Hugging Face dataset repository?

Yes, you can push transformed data to a new repository on the Hugging Face Hub. The Skill supports dataset lifecycle management, allowing you to initialize repositories, configure system prompts, and stream row updates for diverse data formats.

Do I need Python and DuckDB to manage Hugging Face datasets?

Yes, managing Hugging Face datasets requires Python with libraries like DuckDB, Hugging Face Hub, datasets, and Pandas. These dependencies enable initialization, content streaming, and SQL-based querying and transformation within the workflow.

What dataset formats are supported for Hugging Face data transformation?

Data transformation supports diverse dataset types including chat, classification, QA, text completion, and tabular data. This multi-format handling ensures comprehensive dataset workflows for various AI model preparation and manipulation needs.

What is the best way to initialize a new dataset repository on Hugging Face?

The best way to initialize a new dataset repository is through the provided dataset lifecycle management tools. These tools handle repository creation, system prompt configuration, and row update streaming directly integrated with the Hugging Face Hub.

Are there limitations when streaming row updates from Hugging Face datasets?

While streaming row updates is supported for dataset lifecycle management, performance may vary based on dataset size and network conditions. Using SQL via DuckDB for querying and transformation helps mitigate local memory constraints when handling large datasets.