huggingface-skills
🏪Marketplacehuggingface/skills
|Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub
Overview
Hugging Face Skills is an official collection of Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and Hugging Face Hub operations. The skills follow the standardized Agent Skill format and are interoperable with Claude Code, OpenAI Codex, Google Gemini CLI, Cursor, and other coding agent tools.
Key Features
- AI/ML Task Skills — Provides structured guidance for dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub
- Hugging Face CLI Skill — Execute Hub operations including model/dataset downloads, file uploads, repo management, and cloud compute jobs via the hf CLI
- Multi-Agent Compatibility — Works with Claude Code (plugin marketplace), OpenAI Codex (AGENTS.md), Google Gemini CLI (gemini-extension.json), with Cursor and Windsurf integrations on the way
- Standardized Skill Format — Each skill is a self-contained folder with SKILL.md containing YAML frontmatter and actionable guidance for the agent
- Easy Installation — Install via
/plugin marketplace add huggingface/skillsin Claude Code, or usegemini extensions installfor Gemini CLI
Who is this for?
AI/ML practitioners and researchers who use coding agents to accelerate their Hugging Face workflows. It is ideal for teams that want their AI assistants to follow best practices when creating datasets, training models, or publishing research on the Hugging Face Hub.
Add this Marketplace
/plugin marketplace add huggingface/skillsPlugins in this Marketplace
hugging-face-cli
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
hugging-face-datasets
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
hugging-face-evaluation
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
hugging-face-jobs
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
hugging-face-model-trainer
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
hugging-face-paper-publisher
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
hugging-face-tool-builder
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
hugging-face-trackio
Official Hugging Face skills defining AI/ML tasks like dataset creation, model training, and evaluation. Interoperable with Claude Code, OpenAI Codex, Gemini CLI, and Cursor using the standardized Agent Skill format.
More from this repository10
A skill for the Hugging Face Hub CLI (hf), enabling downloading, uploading, and managing models, datasets, spaces, buckets, repos, papers, and jobs on the Hugging Face Hub. Replaces the deprecated huggingface-cli.
Official Hugging Face Gradio skill for building Python web UIs and ML demos — covers the `Interface` high-level wrapper, `Blocks` low-level layout with explicit event wiring, `ChatInterface` for chatbots, core component signatures (Textbox, etc.), streaming inputs/outputs, custom CSS/JS, and sharing your app.
Explores, queries, and extracts data from any Hugging Face dataset via the Dataset Viewer REST API and npx tooling with zero Python dependencies — split/config discovery, row pagination, text search, filtering, SQL via parquetlens, and CLI upload.
Train or fine-tune language models using TRL on Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment.
Hugging Face agent skills for AI/ML tasks like dataset creation, model training, and evaluation, interoperable with major coding agent tools.
Part of the official Hugging Face Skills collection, this skill teaches AI coding agents how to work with Transformers.js for running machine learning models directly in the browser and Node.js environments.
Official Hugging Face skill for training **vision models on HF Jobs cloud GPUs** — fine-tunes object detection (D-FINE, RT-DETR v2, DETR, YOLOS), image classification (timm MobileNetV3/MobileViT/ResNet/ViT/DINOv3 + Transformers classifiers), and SAM/SAM2 segmentation; covers COCO dataset prep (xywh/xyxy auto-detect), Albumentations augmentation, mAP/mAR + accuracy metrics, DiceCE loss, Trackio monitoring, Hub persistence, and mandatory dataset validation before GPU training.
Hugging Face skills for AI/ML tasks including dataset creation, model training, and evaluation. Follows the Agent Skills standard and works with Claude Code, Codex, Gemini CLI, and Cursor.
Hugging Face skill for AI/ML experiment tracking with TrackIO, providing AI coding agents with instructions for logging metrics, parameters, and artifacts during model training and evaluation.
Publish and manage research papers on Hugging Face Hub. Supports creating paper pages, linking papers to models/datasets, claiming authorship, and generating professional markdown-based research articles.