reference-catalog

Maintain and validate SD.Next model reference catalogs in data/reference*.json, including schema consistency, deduplication, link checks, and thumbnail alignment.

Quality

70%

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 ./.github/skills/reference-catalog/SKILL.md

Quality

Content

81%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 body is well-structured, actionable, and includes explicit validation feedback loops appropriate to a batch/destructive catalog-maintenance task. Minor conciseness redundancy and the absence of a complete executable command example for cli/hf-info.py are the main limits.

Suggestions

Consolidate the size-backfill guidance into one section to remove the triple repetition across Core Rules, the Validation Checklist, and Common Failure Modes.

Add one copy-paste-ready example invocation of cli/hf-info.py (including expected output shape) so the size-backfill step is fully executable.

Tighten the fields-check line ('Ensure that all fields are consistent and not null, empty or contain zero values') into a concrete validation rule rather than a restatement.

Dimension	Reasoning	Score
Conciseness	Mostly efficient directive bullets with minimal padding; minor redundancy (size-backfill restated across rules, checklist, and failure modes) and a slightly over-explained fields-check line keep it just below fully lean.	4 / 5
Actionability	Provides concrete tools and formulas (cli/hf-info.py, info.data.size, 'gb = round(mb / 1024, 2)', exact file and category lists), but stops short of a copy-paste-ready full command invocation, leaving a minor gap.	4 / 5
Workflow Clarity	The phased approach plus the 6-step Safe Edit Workflow gives a clear sequence with an explicit re-validation checkpoint ('Re-validate JSON and duplicate checks'), supplying the feedback loop this batch/destructive context requires.	5 / 5
Progressive Disclosure	Self-contained single file with well-organized sections and only one external pointer (core.instructions.md), no nested references; good structure but the ~120-line inlined body is slightly more than a lean overview would warrant, keeping it just below a 5.	4 / 5
	Total	17 / 20 Passed

Description

58%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 clearly states concrete capabilities and is well-scoped to a distinct niche, but it omits an explicit 'Use when' trigger clause and relies on technical jargon rather than natural user phrasing. Adding a trigger clause with everyday synonyms would lift completeness and trigger-term quality.

Suggestions

Append an explicit 'Use when...' clause naming natural trigger phrases (e.g., 'Use when auditing SD.Next model reference files, cleaning duplicate catalog entries, or fixing size/thumbnail metadata').

Add common natural synonyms and file extensions users might actually say (e.g., 'reference JSON files', '.json catalogs') rather than only the glob 'data/reference*.json'.

Lead with the most representative concrete verbs ('audit', 'deduplicate', 'validate') to keep specificity high while tightening phrasing.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions (schema consistency, deduplication, link checks, thumbnail alignment), matching the 'several specific actions; minor gaps' anchor, but not the comprehensive multi-verb coverage of a 5.	4 / 5
Completeness	Has a clear 'what' but lacks any explicit 'Use when...' trigger clause in the description (the trigger guidance lives only in a separate argument-hint), so per the missing-trigger guideline completeness is capped at 3.	3 / 5
Trigger Term Quality	Contains relevant domain terms ('SD.Next model reference catalogs', 'data/reference*.json') but these are technical jargon and file patterns rather than natural user phrases, and common synonyms are missing.	3 / 5
Distinctiveness Conflict Risk	Tied to a specific niche (SD.Next model reference catalogs, named JSON files, thumbnail alignment), making it mostly distinct with only minor overlap risk with closely related catalog/maintenance skills.	4 / 5
	Total	14 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: vladmandic/sdnext
Path: .github/skills/reference-catalog/SKILL.md
Commit: c40fc01

Reviewed: ~17 hours 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.