debug-model

Debug a broken SD.Next or Diffusers model integration. Use when a newly added or ported model fails to load, misdetects, crashes during prompt encoding or sampling, or produces incorrect outputs.

Quality

82%

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

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 delivers a well-sequenced, actionable debugging workflow with strong validation checkpoints and feedback loops, organized into clean sections with a single signaled external reference. Minor gains are available by trimming repetitive symptom phrasing and confirming bundled reference paths.

Suggestions

Tighten the repeated 'Check:' / 'Typical symptoms:' heading pattern across the six failure classes to reduce token overhead without losing the diagnostic structure.

Convert the loader/remap examples in 'Fix The Root Cause' into a short runnable snippet or concrete command to lift actionability to fully executable.

If `.github/instructions/core.instructions.md` is a real bundled reference, surface it as a one-level-deep link in a dedicated references section; otherwise note its optional status to avoid an unverifiable navigation path.

Dimension	Reasoning	Score
Conciseness	The body is largely lean and assumes Claude's competence, organizing failures as checklists rather than re-explaining diffusion concepts, with only minor padding in repeated 'Typical symptoms' phrasing that could be trimmed.	4 / 5
Actionability	It points to specific files, functions, and concrete remediation steps (e.g. 'Add the missing modeldata branch', 'Correct the output path for pixel-space models'), and lists executable validation commands like compileall and ruff, though guidance is instruction-style without runnable code snippets.	4 / 5
Workflow Clarity	A clear outside-in debugging sequence is paired with a 'Minimal Debug Procedure', layered 'Validate In Layers' checkpoints, and a closing 'Validation Checklist' with an explicit feedback loop (validate narrowest layer before proceeding).	5 / 5
Progressive Disclosure	The skill is a self-contained, well-sectioned overview with one clearly signaled external reference (`.github/instructions/core.instructions.md`); structure is good but the body is fairly long and the referenced instruction file is not part of the bundle.	4 / 5
	Total	17 / 20 Passed

Description

83%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, action-oriented, and clearly pairs a 'what' with an explicit 'Use when...' trigger clause, hitting most anchors well. It could be tightened with broader synonym/extension coverage and slightly sharper distinctiveness to reach full marks.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete failure modes ('fails to load, misdetects, crashes during prompt encoding or sampling, or produces incorrect outputs'), but actions described are symptoms rather than the full set of debugging actions performed.	4 / 5
Completeness	Explicitly answers what ('Debug a broken SD.Next or Diffusers model integration') and when via a clear 'Use when...' clause enumerating concrete triggering scenarios.	5 / 5
Trigger Term Quality	Includes natural trigger phrases a user would say ('newly added... fails to load', 'misdetects', 'crashes during prompt encoding or sampling', 'incorrect outputs') with good coverage, though it lacks common synonyms and explicit file-extension or framework keywords beyond SD.Next/Diffusers.	4 / 5
Distinctiveness Conflict Risk	The SD.Next/Diffusers model-port niche is fairly distinct and targeted, but it overlaps somewhat with general model-loading or debugging skills, leaving minor conflict risk with closely related skills.	4 / 5
	Total	17 / 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/debug-model/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.