Name: hugging-face-evaluation
Rating: 60 (1 reviews)
Author: administrakt0r

hugging-face-evaluation

Add and manage evaluation results in Hugging Face model cards. Supports extracting eval tables from README content, importing scores from Artificial Analysis API, and running custom model evaluations with vLLM/lighteval. Works with the model-index metadata format.

Quality

71%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/hugging-face-evaluation/SKILL.md

The canonical home for this skill is hugging-face-evaluation in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

71%Weight 40%Scale 1-5

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

The content is highly actionable with a well-validated core workflow, but it is over-long, redundant across sections, and monolithic — reference material that belongs in separate files is inlined and the referenced scripts are not bundled. Splitting reference material and de-duplicating command examples would materially improve it.

Suggestions

Move the Commands Reference, lighteval task format/availability, inspect-ai task list, hardware recommendations, and Troubleshooting into separate reference files (e.g. references/commands.md, references/tasks.md, references/troubleshooting.md) and link to them from the body.

De-duplicate the CLI examples: the Method 1–4 sections, Commands Reference, and Common Patterns repeat the same commands — keep one canonical set and cross-reference.

Place the version number (1.3.0) under a clearly labeled Version section or remove date/version drift; ensure the referenced scripts/ actually ship in the skill bundle or document where they live.

Dimension	Reasoning	Score
Conciseness	The body is mostly executable commands and code that earn their place, but it is ~650 lines with notable redundancy (Commands Reference and Common Patterns repeat the Method 1–4 commands) and a bare version number '1.3.0' that is time-sensitive and not in a deprecated section.	3 / 5
Actionability	Abundant copy-paste-ready, fully executable commands and code with real model ids and flags covering the common cases (inspect-tables → extract-readme → --create-pr, lighteval/inspect vLLM invocations, HF Jobs submissions).	5 / 5
Workflow Clarity	The main extract-readme workflow is clearly sequenced with explicit validation checkpoints (mandatory get-prs before --create-pr, preview-YAML-before-apply checklist, troubleshooting feedback), though the vLLM custom-eval path (Method 4) has weaker validation checkpoints.	4 / 5
Progressive Disclosure	Section headers give the file structure, but it is a monolithic ~650-line SKILL.md with reference material (Commands Reference, lighteval task format, task lists, hardware table, troubleshooting) inlined rather than split into reference files, and the referenced scripts/ files are not present in any bundle directory.	3 / 5
	Total	15 / 20 Passed

Description

71%Weight 40%Scale 1-5

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, third-person, and clearly scoped to a distinct niche, but it lacks an explicit 'Use when...' trigger clause, which caps its completeness. Adding natural trigger phrases would raise both completeness and trigger-term quality.

Suggestions

Append an explicit 'Use when...' clause, e.g. 'Use when adding evaluation/benchmark results to Hugging Face model cards, importing Artificial Analysis scores, or running vLLM/lighteval evaluations.'

Add common synonyms users would say — 'benchmarks', 'leaderboard scores', 'eval results' — to broaden natural trigger coverage.

Mention .env / HF_TOKEN prerequisites briefly only if relevant to triggering; keep the focus on when-to-use phrasing.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'extracting eval tables from README content', 'importing scores from Artificial Analysis API', 'running custom model evaluations with vLLM/lighteval', plus the model-index metadata format — giving comprehensive coverage of what the skill does.	5 / 5
Completeness	The 'what' is explicit and detailed, but there is no 'Use when...' clause or equivalent explicit trigger guidance, capping completeness at 3 per the rubric guideline.	3 / 5
Trigger Term Quality	Good natural keyword coverage ('Hugging Face model cards', 'evaluation results', 'Artificial Analysis', 'vLLM', 'lighteval', 'model-index') that users would say, though common synonyms like 'benchmarks' or 'leaderboard' are absent.	4 / 5
Distinctiveness Conflict Risk	The HF model-card evaluation niche is clearly scoped with distinct triggers (Artificial Analysis, vLLM, lighteval, model-index), with only minor overlap risk against broader Hugging Face skills.	4 / 5
	Total	16 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (653 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 44 missing	Warning

	Total	13 / 16 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills-claude/skills/hugging-face-evaluation/SKILL.md
Commit: 57c135b

Reviewed: 4 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.