dataset-finder

Search Kaggle, Hugging Face, UCI ML Repository, and Data.gov for datasets.

Updated Feb 8, 2026

Search Tags:#search #preview #datasets #huggingface #kaggle #datacard

Ready to run

npx skills add https://github.com/anisafifi/agent-skills --skill dataset-finder

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dataset-finder
Source: https://github.com/anisafifi/agent-skills/tree/main/dataset-finder
Command: npx skills add https://github.com/anisafifi/agent-skills --skill dataset-finder

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires kaggle, datasets, pandas, huggingface-hub, requests, beautifulsoup4, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill enables users to search, download, preview, and document datasets from multiple repositories (Kaggle, Hugging Face, UCI ML Repository, and Data.gov) in a unified workflow, reducing time spent on data discovery and preparation.

Core Features & Use Cases

Multi-Repository Search across Kaggle, Hugging Face, UCI ML Repository, and Data.gov to locate relevant datasets.
Dataset Download, local organization, and support for batch operations.
Dataset Preview with basic statistics and sample data to assess suitability before use.
Data Card Generation to auto-create documentation for datasets, including schema and usage notes.
List Local Datasets to manage downloaded data assets.

Quick Start

Install OpenClawCLI and dependencies, then run the dataset finder to locate, download, and inspect datasets from multiple sources.

Frequently Asked Questions about dataset-finder

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

FAQPage Schema

How do I search for datasets across Kaggle and Hugging Face at the same time?▼

Cross-repo dataset search queries Kaggle, Hugging Face, UCI ML Repository, and Data.gov simultaneously. It uses the Kaggle API and Hugging Face Hub to locate relevant datasets across multiple repositories within a single unified workflow.

Can I preview dataset statistics before downloading from Hugging Face or Kaggle?▼

Dataset preview displays basic statistics and sample data to assess suitability before use. It leverages pandas to inspect downloaded files, helping you evaluate datasets from Hugging Face or Kaggle without committing to full integration.

How do I automatically generate data cards for machine learning datasets?▼

Automatic data card generation creates documentation for datasets, including schema and usage notes. It extracts metadata from downloaded files using pandas and BeautifulSoup to produce structured documentation for your machine learning projects.

Do I need API credentials to search and download datasets from Kaggle and Hugging Face?▼

Kaggle API and Hugging Face Hub dependencies require proper authentication credentials configured in your environment. You must set up these API tokens before the skill can execute search, download, or batch operations across the supported repositories.

What's the best way to organize downloaded datasets from multiple repositories locally?▼

Local dataset organization manages downloaded data assets from Kaggle, Hugging Face, UCI ML Repository, and Data.gov. The skill lists local datasets and supports batch operations to keep files structured for downstream machine learning tasks.

Why use a cross-repo dataset finder instead of searching individual platforms separately?▼

A cross-repo dataset finder reduces time spent on data discovery and preparation by searching multiple repositories in one workflow. It unifies search, download, preview, and documentation tasks, eliminating the need to navigate each platform individually.