skills.rest
cpich3g
cpich3gCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

hugging-face-datasets

Automate Hugging Face dataset creation, configuration, and SQL-based management via DuckDB.

Updated Feb 25, 2026
Search Tags:#templates#sql#datasets#huggingface#hub#dataset-management
One-click install
npx skills add https://github.com/cpich3g/ai-cost-optimiser --skill hugging-face-datasets-cpich3g
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hugging-face-datasets
Source: https://github.com/cpich3g/ai-cost-optimiser/tree/main/.github/skills/hugging-face-datasets
Command: npx skills add https://github.com/cpich3g/ai-cost-optimiser --skill hugging-face-datasets-cpich3g

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Create and manage Hugging Face datasets across repos with config, system prompts, streaming updates, and SQL-based workflows.

Core Features & Use Cases

  • Initialize and configure dataset repos on the Hugging Face Hub, including system prompts and metadata.
  • Stream updates with JSONL data and perform SQL-based querying and transformations via DuckDB using hf:// paths.
  • Push results back to the Hub or push derived datasets to new repos for downstream ML pipelines and research datasets.

Quick Start

Initialize a HF dataset repo, apply a system prompt, and add an initial batch of examples using the provided scripts.

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 Hugging Face datasets with SQL?▼

Query Hugging Face datasets with SQL by using DuckDB-backed SQL queries on hf:// paths. This allows you to perform data wrangling and transformations directly on Hub datasets without manual downloads.

How do I automate creating and managing Hugging Face dataset repos?▼

Automate creating and managing Hugging Face dataset repos by initializing them with configuration, system prompts, and metadata. You can then stream JSONL updates and push derived datasets to new repos for downstream pipelines.

Can I push transformed dataset results back to the Hugging Face Hub?▼

You can push transformed dataset results back to the Hugging Face Hub after applying SQL-based wrangling. Derived datasets can also be pushed to new repos to support downstream ML pipelines and research.

Do I need a Hugging Face token to manage datasets across multiple repos?▼

A Hugging Face token is required to authenticate and manage datasets across multiple repos. The token enables initializing repos, streaming updates, and pushing results back to the Hub securely.

What is the best way to perform batch updates on Hugging Face datasets?▼

The best way to perform batch updates on Hugging Face datasets is through template-driven validation and streaming JSONL data. This supports ongoing configuration and automated management across projects.

Does DuckDB work with hf:// paths for dataset transformations?▼

DuckDB works directly with hf:// paths to execute SQL-based querying and transformations on Hugging Face datasets. This integration enables efficient data wrangling on Hub repos before pushing results back.

Related Skills

Explore More Data & Analytics Skills→
huggingfacehuggingface
10.9k

hf-mcp

Search and retrieve Hugging Face Hub models, datasets, and Spaces via MCP server.

Official
Advanced
patchy631patchy631
36.9k

hugging-face-datasets

Manage Hugging Face Hub datasets with SQL queries via DuckDB.

Community
Intermediate
crossxwillcrossxwill
5

duckdb-parquet-lab-workflow

Query Parquet data with DuckDB and convert results to pandas DataFrames.

Community
Intermediate
luokai25luokai25

hugging-face-datasets

Manage Hugging Face Hub dataset lifecycles with SQL queries and transformations.

Community
Advanced