huggingfaceOfficial·32 Agent Skills Included

skills

Train, evaluate, and deploy AI models on Hugging Face

Teaches AI agents to run the full Hugging Face workflow: training LLMs and vision models on cloud GPUs, evaluating models, building Gradio demo apps, and deploying to SageMaker. Removes guesswork around GPU memory sizing, dataset formats, IAM roles, and container selection that cause most ML job failures. Includes ready-made scripts and checklists so agents complete training and deployment tasks correctly on the first attempt.

10,996|738

↗

npx skills add huggingface/skills --all -g -y

Available:

Tells the AI agent which Hugging Face skill to load for each user request, routing tasks like model training, evaluation, Spaces deployment, and SageMaker hosting to the correct SKILL.md file.

All Skills in This Repository (32)

Pure Emerald Level Indicators

📦 In Repo

huggingface

huggingface-papers

Fetch and process AI research papers from the Hugging Face Papers API.

huggingface-llm-trainer

Train language and vision models with TRL methods on Hugging Face Jobs.

huggingface-lora-space-builder

Generate Gradio web apps for AI models and publish them on Hugging Face Spaces.

transformers-js

Run Hugging Face machine learning models in JavaScript and TypeScript applications.

huggingface-community-evals

Evaluate Hugging Face Hub models locally with inspect-ai and lighteval.

hf-cloud-sagemaker-iam-preflight

Verify or create AWS SageMaker IAM execution roles with AWS CLI.

huggingface-zerogpu

Manage AI/ML tasks on Hugging Face Spaces with ZeroGPU constraints.

hf-cloud-sagemaker-deployment-planner

Automates AI model deployment to Amazon SageMaker with real-time, serverless, async options.

huggingface-spaces

Create and deploy ML apps on Hugging Face Spaces with Gradio, Docker, or Static.

huggingface-paper-publisher

Automate publishing and managing research papers on the Hugging Face Hub.

hf-cloud-serving-image-selection

Select serving containers and retrieve image URIs for SageMaker deployments.

train-sentence-transformers

Automate training and fine-tuning of sentence-transformers models for AI tasks.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install Hugging Face Skills?▼

Run `npx skills add huggingface/skills --all -g -y` in your terminal to install the full skill suite globally for your AI agent.

How to train an LLM without a local GPU?▼

The huggingface-llm-trainer skill lets your agent submit TRL training jobs (SFT, DPO, GRPO) to Hugging Face Jobs cloud GPUs, with automatic cost estimation and Hub saving.

How to deploy a model to AWS SageMaker with AI?▼

The SageMaker skills guide your agent through the full deployment: discovering your AWS context, checking IAM roles, picking the right container, and creating a production endpoint with autoscaling and alarms.

Does Hugging Face Skills work with Claude Code and Cursor?▼

Yes. All skills follow the standard SKILL.md format and work with Claude Code, Codex, Gemini CLI, and Cursor.

Can I run models locally with these skills?▼

Yes. The huggingface-local-models skill helps your agent find GGUF models, pick the right quantization, and serve them locally with llama.cpp on CPU, Mac, or GPU.