Hugging Face Skills

Hugging FaceResearch skill collectionsListed

About this tool

Official Hugging Face agent skills. Manage models, datasets and Spaces through hf CLI, view and publish paper pages, fine-tune with TRL or Unsloth on HF Jobs cloud GPUs, evaluate with lighteval, track experiments with Trackio and build Gradio demos. Install selected skills in Claude Code, Codex, Gemini CLI or Cursor.

  • Feature 1: hf-cli manages Hub models, datasets, Spaces and Jobs.
  • Feature 2: huggingface-datasets searches and filters datasets and downloads parquet files.
  • Feature 3: huggingface-papers and huggingface-paper-publisher read paper metadata and publish paper pages.
  • Feature 4: huggingface-llm-trainer, trl-training and huggingface-vision-trainer support model training.
  • Feature 5: huggingface-community-evals and huggingface-trackio provide local evaluation and experiment tracking.

Resource information

Maintainer
Hugging Face; official project.
Type
Skill collection
Category
Research skill collections
Runs on
Agent Skills
Who it's for
Researchers and groups using Hugging Face Hub to find datasets and models, fine-tune and evaluate, track experiments, and publish papers or demo Spaces.
License at time of listing
Apache-2.0
Source
GitHub