Query and transform Hugging Face datasets with DuckDB SQL.
npx skills add https://github.com/huggingface/skills --skill hugging-face-datasets
Please help me install this Agent Skill. Skill: hugging-face-datasets Source: https://github.com/huggingface/skills/tree/main/skills/hugging-face-datasets Command: npx skills add https://github.com/huggingface/skills --skill hugging-face-datasets
Hugging Face datasets often require manual setup for repo creation, configuration, and data processing. This Skill provides an end-to-end workflow to initialize, configure, and edit datasets, plus SQL-based discovery, transformation, and export capabilities.
Use uv run scripts/dataset_manager.py init to create a new dataset, then uv run scripts/dataset_manager.py quick_setup --template chat --repo_id "your-username/your-dataset" to bootstrap a dataset with chat templates. Then run uv run scripts/sql_manager.py query --dataset "your-username/your-dataset" --sql "SELECT * FROM data" to inspect.
High-intent search queries and answers about installing and using this skill.
You can query Hugging Face datasets using SQL by leveraging DuckDB integration. The skill supports SQL-based data manipulation, allowing you to describe schemas, sample data, and join datasets directly across multiple splits.
The best way to initialize a Hugging Face dataset is using the provided Python scripts to create repos and bootstrap configurations. You can apply templates, like chat templates, and manage content across multiple splits programmatically.
Yes, you can export Hugging Face dataset query results to Parquet or JSONL. The skill supports SQL-based transformation and export capabilities, enabling you to push processed results to new datasets on the Hugging Face Hub.
Yes, you need an HF_TOKEN environment variable set up to manage private Hugging Face datasets. The skill relies on Hugging Face Hub APIs and requires this token for authentication during initialization, querying, and pushing results.
Yes, the DuckDB integration supports joining multiple Hugging Face datasets. You can use standard SQL commands to query, join, and transform data across different dataset splits and configurations seamlessly.
LAF-USInstall Codex skills from curated lists or GitHub repositories.
Himanshu040604Query and transform Hugging Face datasets using SQL via DuckDB.
POWERFULMOVESCreate and manage Hugging Face Hub datasets with streaming support.
luokai25Manage Hugging Face Hub dataset lifecycles with SQL queries and transformations.
