imoonkey

Hugging Face

```

imoonkey 56 3 Updated 4mo ago

Resources

6
GitHub

Install

npx skillscat add imoonkey/openweb/src-sites-huggingface

Install via the SkillsCat registry.

About this skill

This skill provides access to the Hugging Face Hub's public REST API, enabling agents to search and retrieve information about machine learning models, datasets, and demo applications. It solves the problem of programmatically discovering and researching AI resources without manual web browsing. Developers should use it when building agents that need to find relevant models or datasets based on search queries or explore details like model architectures, dataset descriptions, and associated demo apps.

SKILL.md

Hugging Face

Overview

AI model and dataset hub. Public REST API for searching and browsing ML models, datasets, and Spaces (demo apps).

Workflows

Find an ML model

  1. searchModels(search)id (owner/name format)
  2. getModel(owner, name) → pipeline_tag, downloads, tags, cardData, siblings

Explore datasets

  1. searchDatasets(search)id (owner/name format)
  2. getDataset(owner, name) → cardData, description, citation, siblings

Discover demo apps

  1. getSpaces(search)id, sdk, runtime.stage

Research a model and its data

  1. searchModels(search)id (owner/name)
  2. getModel(owner, name)cardData.datasets (referenced dataset names)
  3. searchDatasets(search) → find referenced dataset → id
  4. getDataset(owner, name) → description, citation

Operations

Operation Intent Key Input Key Output Notes
searchModels find ML models search id, author, downloads, pipeline_tag, tags sortable by downloads/likes/trending
getModel model details owner, name <- searchModels id, pipeline_tag, downloads, tags, cardData, siblings includes file list and related Spaces
searchDatasets find datasets search id, author, downloads, tags sortable by downloads/likes/trending
getDataset dataset details owner, name <- searchDatasets id, downloads, tags, cardData, description, citation includes file list
getSpaces browse demo apps search id, author, likes, sdk, runtime sortable by likes/trending

Quick Start

# Search for models
openweb huggingface exec searchModels '{"search": "text-generation", "limit": 5}'

# Get model details
openweb huggingface exec getModel '{"owner": "meta-llama", "name": "Llama-2-7b"}'

# Search for datasets
openweb huggingface exec searchDatasets '{"search": "sentiment", "limit": 5}'

# Get dataset details
openweb huggingface exec getDataset '{"owner": "stanfordnlp", "name": "imdb"}'

# Browse Spaces
openweb huggingface exec getSpaces '{"search": "chatbot", "limit": 5}'