diffusers-code

Create or edit code that is compliant with Hugging Face diffusers conventions, including models, pipelines, schedulers, tests, docs, and PR preparation targeting diffusers.

Quality

56%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.github/skills/diffusers-code/SKILL.md

Quality

Content

62%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 a well-organized, lean rule set with a clear sequenced workflow and validation commands, but it offers little executable code and relies on a single non-bundled reference. Adding runnable diffusers snippets and either creating or clearly signaling bundled reference files would lift actionability and progressive disclosure.

Suggestions

Add 1-2 short executable code snippets (e.g. a register_to_config model stub and a @torch.no_grad() __call__ signature) to make guidance copy-paste ready.

Make the validation feedback loop explicit in the Change Workflow (e.g. 'If focused tests fail: fix and re-run before proceeding').

Move the long rule lists (Models/Attention/Pipelines/Schedulers) into a bundled reference file and link to it, or confirm core.instructions.md exists in the bundle and reference it more explicitly.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean bullet-style guidance that assumes Claude's knowledge of diffusers; minor phrases like 'before proceeding' and a few restated objectives could be trimmed, but it largely avoids explaining concepts Claude already knows.	4 / 5
Actionability	It gives concrete rules (register_to_config, @torch.no_grad(), make style, make fix-copies) but lacks executable code examples or copy-paste-ready snippets; guidance is mostly imperative prose rather than runnable code.	3 / 5
Workflow Clarity	The Change Workflow is clearly sequenced (Gather, Plan, Implement, Validate, Polish) with validation commands and a re-run step; minor gap is that validation feedback loops are implicit rather than an explicit 'if fails, fix and re-run' loop.	4 / 5
Progressive Disclosure	Content is well sectioned but the only external reference is .github/instructions/core.instructions.md, which is not a bundled file here, and detailed rules that could live in reference files are inlined; structure is decent but references are not clearly signaled against an actual bundle.	3 / 5
	Total	14 / 20 Passed

Description

51%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 conveys what the skill does and is well-scoped to diffusers, but it omits any explicit 'Use when...' trigger guidance and relies on technical jargon rather than natural user phrasing. Adding concrete trigger conditions would substantially raise completeness and trigger-term quality.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when implementing or reviewing code for the Hugging Face diffusers library or preparing a diffusers PR.'

Include natural user phrasings and synonyms such as 'diffusers pipeline', 'add a scheduler', 'write a diffusers model', and 'diffusers PR' to improve trigger-term quality.

Mention 'loaders' and 'processors' explicitly if they are in scope, since the body covers them but the description omits them.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Create or edit code', models, pipelines, schedulers, tests, docs, and PR preparation) with minor gaps in coverage such as loaders and processors being only implied.	4 / 5
Completeness	The 'what' is clear but there is no explicit 'when' / 'Use when...' clause, which the guidelines cap at 3; the argument-hint gives context but the description itself never states trigger conditions.	2 / 5
Trigger Term Quality	Includes domain terms like 'models', 'pipelines', 'schedulers', and 'PR preparation' but lacks the natural phrases a user would actually say ('write a diffusers model', 'add a scheduler') and offers no synonyms.	3 / 5
Distinctiveness Conflict Risk	Scoped tightly to 'Hugging Face diffusers conventions' targeting the diffusers repository, making it distinct from generic code skills with only minor overlap with broader HF/library skills.	4 / 5
	Total	13 / 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/diffusers-code/SKILL.md
Commit: c40fc01

Reviewed: ~18 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.