1. Habit
  2. Skills
  3. Hugging Face Datasets skill
huggingface · 2K installs

Hugging Face Datasets skill
Runs by itself, every Monday at nine.

Spot dataset changes every week.

Install the Hugging Face Datasets skill and make it a habit: follow https://gethabit.dev/skills/huggingface/huggingface-datasets.md

  1. 1Copy
  2. 2Paste into Claude Code, Codex or Cursor
  3. 3It runs by itself

2K installs11K starsApache-2.0 licenceSource on GitHub ↗

Hugging Face Datasets

by huggingface · 2K installs

The skill

The hugging face datasets skill lets an agent explore datasets through the Dataset Viewer API: splits, row pages, text search, filters, parquet links, size and statistics, plus upload flows.

Only when you ask.

The habit

Spot dataset changes every week.

↻ Runs by itselfevery Monday at nine

How the habit works.
You don't have to.

Three questionsWhen does it start? How does it check its work? What does it remember?

01

It starts by itself

every Monday at nine. Runs in Claude Code routine or cron.

  • ↻ every Monday at nine
  • 0 9 * * 1

02

It checks its work

dataset valid, rows counted. If a check fails, it tells you.

  • ✓/is-valid reports the dataset as valid
  • ✓/splits returns at least one config and split
  • ✓/size returns a row total above zero

03

It remembers

dataset size log. The next run reads it first: last row totals per dataset and split.

loops/hf-dataset-health-loop/run-log.md

One run, step by step.
Then it stops.

Stop ruleStops after at most 10 datasets per run and one retry each.

  1. 01

    Read the log

    Your agent loads the dataset list and last week's totals.

  2. 02

    Validate datasets

    It calls /is-valid for each dataset.

  3. 03

    Resolve splits

    It calls /splits to get each config and split.

  4. 04

    Read size and stats

    It calls /size and /statistics, with a bearer token for gated or private datasets.

  5. 05

    Preview rows

    It calls /first-rows to confirm rows still come back.

  6. 06

    Compare and log

    It compares totals with the last run and writes changes to the log.

What it may do.
Nothing more.

Safe by defaultIt writes only in its own folder and asks you before anything goes out.

Is it for you?
Use it when it fits.

Quick checkRead both lists. If the left one sounds like you, the Hugging Face Datasets skill helps.

Use it when

  • ✓ You explore a Hub dataset
  • ✓ You need parquet links or row counts
  • ✓ You search or filter dataset rows

Skip it when

  • – Your data is not on the Hub
  • – You need SQL over parquet

Questions

Hugging Face Datasets skill, answered.

Does it change my datasets?

No. The loop only uses read-only Dataset Viewer calls.

Can it check private datasets?

Yes, with an Authorization bearer header that holds your HF_TOKEN.

How do I install the Hugging Face Datasets skill?

Copy the prompt at the top into your agent. It installs the skill from huggingface/skills and sets up the habit. Or run npx skills add huggingface/skills --skill huggingface-datasets.

Is the Hugging Face Datasets skill free?

Yes. It is published under Apache-2.0. Habit is free too.

Hugging Face Datasets. Now as a habit.

Install the Hugging Face Datasets skill and make it a habit: follow https://gethabit.dev/skills/huggingface/huggingface-datasets.md

  • Free
  • No account
  • Undo in one line

All huggingface skills →More ops skills →

Works in

  • Claude Code
  • Codex
  • Cursor
  • ChatGPT
  • Claude app

Made by huggingface, published under Apache-2.0. Habit links to the original and never changes it. Is this your skill?

Copying didn't work here. Select the line and copy it by hand.