Agent Skill : hf-jobs

Run any workload on Hugging Face Jobs.

Use this skill when you want to run GPU/CPU workloads (batch inference, synthetic data generation, dataset stats, experiments) on Hugging Face Jobs, with correct token handling and result persistence back to the Hub.

Skill Name: hf-jobs

Main Documentation: hf-jobs/SKILL.md

Scripts Directory: hf-jobs/scripts/

References Directory: hf-jobs/references/

Overview

This skill focuses on running real workloads via Hugging Face Jobs. It includes ready-to-run UV scripts and guides for authentication (HF tokens), secrets vs env vars, timeouts, hardware selection, and pushing results to the Hub.

Core Documentation

SKILL.md
hf-jobs/SKILL.md

Complete skill documentation (how to submit jobs, tokens/secrets, timeouts, persistence, and how to use the bundled scripts)

References

token_usage.md
hf-jobs/references/token_usage.md

Token best practices: secrets vs env, permissions, common errors (401/403), and secure patterns
hub_saving.md
hf-jobs/references/hub_saving.md

How to persist results: push datasets/models/files to the Hub (ephemeral job filesystem)
hardware_guide.md
hf-jobs/references/hardware_guide.md

Flavor selection guidance for CPU/GPU/TPU workloads
troubleshooting.md
hf-jobs/references/troubleshooting.md

Common failure modes (timeouts, missing deps, OOM, auth) and fixes

Scripts

generate-responses.py
hf-jobs/scripts/generate-responses.py

vLLM batch generation: load prompts/messages from a dataset, generate responses, push dataset + card to Hub
cot-self-instruct.py
hf-jobs/scripts/cot-self-instruct.py

CoT Self-Instruct synthetic data generation (reasoning/instruction) + optional filtering, pushes dataset + card
finepdfs-stats.py
hf-jobs/scripts/finepdfs-stats.py

Polars streaming stats over Hub parquet (finepdfs-edu); optional upload of computed stats to a dataset repo