🤗

Huggingface Hub

Search, download, and upload models & datasets, manage repos, query datasets with SQL, deploy inference endpoints, and manage Spaces on Hugging Face Hub.

作者 v1.0.0
AI & LLMs
正在連線至 VM...
正在連線至 VM...
npx clawhub@latest install huggingface-hub
1目前安裝數
v1.0.0版本

The Hugging Face Hub skill connects your AI assistant to the full Hugging Face ecosystem. It lets you search for models and datasets, download them locally, push your own work to the Hub, query dataset splits with SQL, deploy and manage inference endpoints, and control Spaces and storage buckets — all through natural conversation.

運作原理

This skill leverages the hf CLI under the hood. When you ask your assistant to find a model, pull a dataset, or deploy an endpoint, it translates your request into the appropriate hf CLI commands, executes them in your environment, and returns structured results. Authentication is handled through your existing HF access token.

核心功能

Model Search
Find models by name, task, size, or license across the entire Hugging Face Hub catalogue.
Dataset Download
Pull datasets or individual splits directly to a local path of your choice.
Model & Dataset Upload
Push fine-tuned models, curated datasets, or updated files to your HF account or organisation repos.
Repo & File Management
Create, delete, and manage repositories, files, and revisions on the Hub.
SQL Dataset Queries
Query dataset splits using SQL syntax without downloading the full dataset first.
Inference Endpoint Deployment
Deploy models to dedicated inference endpoints and manage their lifecycle.
Spaces Management
Create, update, and monitor Hugging Face Spaces and associated storage buckets.

系統需求

hf CLI
Must be installed and accessible in your terminal environment.
Hugging Face Account
A valid HF account is required for upload and management operations.
Access Token
You must be authenticated via hf login or an equivalent token configuration.
Gated Model Awareness
Some models require explicit access approval on the Hub; license terms must be respected.

使用情境

Research Prototyping
Quickly locate and download a small vision-language model under 4B parameters for local experimentation.
Dataset Ingestion
Pull a specific training split of a large dataset to a local directory for fine-tuning pipelines.
Model Publishing
Push a newly fine-tuned checkpoint to your HF profile and share it with the community.
Data Exploration
Run SQL queries against a remote dataset split to inspect schema and sample rows before committing to a full download.
Production Deployment
Deploy a model to a dedicated inference endpoint and monitor its status without leaving your assistant.
Space Automation
Programmatically update or restart a Hugging Face Space as part of a CI/CD workflow.

安裝方式

1
Run in your terminal
npx clawhub@latest install huggingface-hub
or
2
Click the Install button at the top of this page for one-click setup

評價

0 則評價

登入後撰寫評價

尚無評價。來分享你的使用體驗吧！

Huggingface Hub Claude Code Skill | MyClaw.ai