Language: 日本語English

huggingface-skills Plugin

Category
Development
Topics
AI Agents & AI App Development · Claude Code Customization & Workflow
First seen
2026-07-09
Last confirmed
2026-07-13
Explanation last updated
2026-07-12

The explanation below is AI-generated. Please verify it against the sources.

huggingface-skills is a plugin (GitHub repository huggingface/skills) that packages a collection of "Agent Skills" — self-contained instruction folders (SKILL.md plus scripts/resources) that teach AI coding agents how to perform Hugging Face AI/ML tasks such as dataset creation, model training, and evaluation. According to the README, the client plugin marketplaces expose only the hf-cli skill as the recommended bootstrap install, while additional workflow skills (over 20 are listed) can be added on demand with hf skills add . It is compatible with Claude Code, Codex, Gemini CLI, and Cursor, and is licensed under Apache-2.0.

Overview

Not a SaaS product itself; it is an open-source GitHub repository (huggingface/skills) distributing Agent Skills — standardized instruction bundles that coding agents load to interact with the Hugging Face Hub and related ML workflows.

What you can do with huggingface-skills

  • Install hf-cli as the first skill to give an agent access to the Hugging Face Hub (search models, manage datasets/buckets, launch Spaces, run jobs), per the README
  • Add further workflow skills on demand with hf skills add (e.g. huggingface-llm-trainer, huggingface-vision-trainer, huggingface-spaces, huggingface-datasets, huggingface-papers, train-sentence-transformers, transformers-js, trl-training, and others listed in the README)
  • Register the repository as a plugin marketplace and install skills in Claude Code, Codex, Gemini CLI, or Cursor, as described in the Installation section
  • Instruct a coding agent directly (e.g. "Use the HF LLM trainer skill to estimate the GPU memory needed for a 70B model run") so it loads the matching SKILL.md instructions and helper scripts
  • Copy an existing skill folder to create and contribute a new custom skill, following the frontmatter format shown in the README
  • Use AWS/SageMaker-oriented skills (hf-cloud-aws-context-discovery, hf-cloud-sagemaker-deployment-planner, hf-cloud-sagemaker-iam-preflight, hf-cloud-sagemaker-production-defaults, hf-cloud-serving-image-selection) for planning and deploying models to Amazon SageMaker AI

Sources

Original description (English)

Build, train, evaluate, and use open source AI models, datasets, and spaces.

History of huggingface-skills

Back to list