huggingface-hub

Manage Hugging Face Hub repositories, datasets, and inference endpoints via the hf CLI.

1|Updated Apr 24, 2026
One-click install
npx skills add https://github.com/automatedigital/spark --skill huggingface-hub-automatedigital
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: huggingface-hub
Source: https://github.com/automatedigital/spark/tree/main/skills/mlops/huggingface-hub
Command: npx skills add https://github.com/automatedigital/spark --skill huggingface-hub-automatedigital

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Data scientists and ML engineers waste time switching between the Hugging Face web dashboard and their local terminal to manage models, datasets, and deployment infrastructure, slowing down iteration and MLOps workflows.

Core Features & Use Cases

  • Repository Management: Create, duplicate, move, and delete model, dataset, and Space repositories directly from the terminal, no web UI required.
  • Asset Operations: Search, download, upload, and run SQL queries against dataset parquet files, plus manage model metadata and files.
  • Infrastructure & Collaboration: Deploy and scale inference endpoints, manage Spaces, handle discussions and pull requests, and automate workflows with webhooks.
  • Use Case: A data scientist can use this skill to download a public sentiment analysis model, upload a fine-tuned version to their private repository, and deploy a test inference endpoint, all without leaving their terminal.

Quick Start

Use the huggingface-hub skill to download the 'distilbert-base-uncased-finetuned-sst-2-english' sentiment analysis model and list all files in its repository.

Frequently Asked Questions about huggingface-hub

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

FAQPage Schema
How do I manage Hugging Face Hub repositories from the command line?

You can manage Hugging Face Hub repositories from the command line by using the hf CLI tool to create, duplicate, move, and delete model, dataset, and Space repositories without needing the web UI.

Can I run SQL queries against Hugging Face datasets without downloading them?

Yes, you can run SQL queries against dataset parquet files stored on the Hugging Face Hub directly from your terminal using the hf CLI tool to query and curate datasets.

How do I deploy Hugging Face inference endpoints via CLI?

You can deploy and scale Hugging Face inference endpoints via CLI by using the hf tool to orchestrate compute resources and manage infrastructure for your MLOps workflows.

What is the best way to automate Hugging Face Hub workflows?

The best way to automate Hugging Face Hub workflows is by using the hf CLI tool to handle file transfers, repository operations, and webhook automation directly from your terminal environment.

Does the hf CLI tool support managing model versions and collaborative pull requests?

Yes, the hf CLI tool supports collaborative Hub contributions by allowing you to handle discussions, manage pull requests, and perform model versioning directly from the terminal.

Can I upload a fine-tuned model to a private Hugging Face repository from the terminal?

Yes, you can upload a fine-tuned model to a private Hugging Face repository from the terminal by using the hf CLI tool to manage asset operations, file transfers, and model metadata.