Do I need a paid Hugging Face account to use this skill?

Yes, Hugging Face Jobs requires a Pro, Team, or Enterprise plan to access the managed cloud infrastructure.

What hardware options are available for my jobs?

You can specify different hardware 'flavors' such as cpu-basic, a10g-small, a10g-large, and various TPU configurations depending on your plan.

How do I handle authentication for remote jobs?

The skill uses the HF_TOKEN secret. When using the hf_jobs MCP tool, passing '$HF_TOKEN' as a secret value will automatically replace it with your local authenticated token.

Can I run scripts with custom dependencies?

Yes, you can use PEP 723 inline dependency headers in your Python scripts, or specify a list of requirements via the dependencies parameter in the hf_jobs tool.

How long can a single job run?

By default, jobs have a 30-minute timeout, but this can be adjusted in the configuration for long-running tasks like model training.

Hugging Face Jobs

Name: Hugging Face Jobs
Author: sickn33

bysickn33

•

31,722

•

Ciencia de Datos y ML

Executes intensive Python workloads and machine learning tasks on managed Hugging Face cloud infrastructure including GPUs and TPUs.

This skill enables Claude to provision and manage remote compute resources on the Hugging Face Hub, allowing you to run heavy workloads without any local setup. It leverages the hf_jobs tool to execute Python scripts via UV (PEP 723) or custom Docker images, providing access to scalable CPUs, GPUs, and TPUs. Ideal for data processing at scale, batch inference, model fine-tuning, and long-running experiments, the skill manages authentication via HF_TOKEN and ensures results are persisted directly back to the Hugging Face Hub.

Características Principales

01Support for custom Docker images and specific Python versions

02Managed cloud compute (CPU, GPU, and TPU flavors)

0331,722 GitHub stars

04Automatic result persistence to the Hugging Face Hub

05Inline UV script execution with PEP 723 dependency management

06Secure secret and environment variable handling (HF_TOKEN support)

Casos de Uso

01Fine-tuning large language models or running TRL-based training workflows

02Executing scheduled or batch inference tasks across thousands of samples

03Processing and transforming massive datasets that exceed local hardware limits

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add sickn33/antigravity-awesome-skills hugging-face-jobs

For use in Claude.ai and ChatGPT

Características Principales

01Support for custom Docker images and specific Python versions

02Managed cloud compute (CPU, GPU, and TPU flavors)

0331,722 GitHub stars

04Automatic result persistence to the Hugging Face Hub

05Inline UV script execution with PEP 723 dependency management

06Secure secret and environment variable handling (HF_TOKEN support)

Casos de Uso

01Fine-tuning large language models or running TRL-based training workflows

02Executing scheduled or batch inference tasks across thousands of samples

03Processing and transforming massive datasets that exceed local hardware limits

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add sickn33/antigravity-awesome-skills hugging-face-jobs

For use in Claude.ai and ChatGPT