huggingface-datasets

Use this skill for Hugging Face Dataset Viewer API workflows that fetch subset/split metadata, paginate rows, search text, apply filters, download parquet URLs, and read size or statistics.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

100%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a tight, highly actionable API reference with executable curl/jq/npx examples, a clear sequenced core workflow, and verification on the write path. It is well organized and self-contained with no bloated concept explanations.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — it lists endpoints with parameters and executable examples without explaining what a dataset or API is, and every section earns its tokens.	3 / 3
Actionability	Guidance is fully executable and copy-paste ready: real curl calls (stanfordnlp/imdb), a jq pipeline to derive parquet alias paths, and concrete npx parquetlens SQL commands with CSV/JSON/Parquet export variants.	3 / 3
Workflow Clarity	The 'Core workflow' is a clearly numbered 6-step sequence, and the destructive write path (dataset upload) includes an explicit verification step ('Verify shards appear in Dataset Viewer' via /parquet) plus a post-upload /parquet discovery step.	3 / 3
Progressive Disclosure	The skill is a single self-contained file with no bundle references, organized into clear sections (Core workflow, Defaults, Dataset Viewer, Querying, Creating/Uploading) that make navigation easy without nested references.	3 / 3
	Total	12 / 12 Passed

Description

67%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and distinctive, clearly scoping a narrow Hugging Face Dataset Viewer API niche with a concrete action list. Its main weakness is the 'when' trigger, which is domain-implied rather than an explicit user-scenario clause, and somewhat technical phrasing.

Suggestions

Add an explicit 'Use when the user wants to browse, search, filter, or download rows from a Hugging Face dataset' clause to convert the implied trigger into a concrete user-scenario trigger.

Soften jargon like 'Dataset Viewer API workflows that fetch subset/split metadata' toward natural phrasings a user would say (e.g., 'browse Hugging Face datasets', 'download dataset rows as parquet').

Include common keyword variants (e.g., 'HF', 'Hugging Face datasets') so the trigger matches how users actually refer to the service.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'fetch subset/split metadata, paginate rows, search text, apply filters, download parquet URLs, and read size or statistics' — matching the anchor for enumerating specific capabilities rather than vague language.	3 / 3
Completeness	The 'what' is clearly answered by the concrete action list, but the 'when' is only domain-implied ('for Hugging Face Dataset Viewer API workflows') rather than an explicit user-scenario trigger such as 'Use when the user needs to browse or download Hugging Face datasets'.	2 / 3
Trigger Term Quality	Relevant domain keywords appear ('Hugging Face', 'datasets', 'rows', 'search', 'filter', 'parquet') but the framing 'Dataset Viewer API workflows that fetch...' is technical jargon and common user phrasings like 'HF', 'browse', or 'download dataset rows' are missing.	2 / 3
Distinctiveness Conflict Risk	The niche is narrow and distinctive ('Hugging Face Dataset Viewer API', 'subset/split', 'parquet'), making it unlikely to trigger for unrelated skills.	3 / 3
	Total	10 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: openai/plugins
Path: plugins/hugging-face/skills/datasets/SKILL.md
Commit: 11c74d6

Reviewed: 17 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.