seedance-troubleshoot

This skill should be used when a Seedance 2.0 output is blurry, jittery, off-prompt, morphing, blocked, visually generic, unstable, desynced, inconsistent, or otherwise fails and needs root-cause diagnosis.

Quality

91%

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

92%

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

A tight, highly actionable diagnostic skill body with a clear repair workflow; its single real weakness is that the many [ref:...] pointers advertise reference files that are not present in the bundle, undermining the progressive-disclosure structure.

Suggestions

Ship the referenced bundle files (references/model-mechanics.md, failure-atlas.md, delivery-qc.md, etc.) so the [ref:...] pointers resolve, or remove pointers to files that do not exist.

Convert [ref:name] tokens into standard markdown links (e.g., [model-mechanics](references/model-mechanics.md)) so the references are clickable and verifiable rather than custom notation.

Dimension	Reasoning	Score
Conciseness	The body is information-dense and assumes Claude's competence — a compact diagnostic table, a fill-in retry template, and terse process steps with no padding explaining what Seedance or video generation is; the only soft spot is the brief editorializing 'Intent' section, which is short enough not to disqualify the lean anchor.	3 / 3
Actionability	Guidance is concrete and executable throughout: a per-symptom diagnostic table with specific first repairs, a slot-fillable 'Conservative Retry Pattern' template, and explicit escalation actions — fully actionable for an instruction-only skill.	3 / 3
Workflow Clarity	The 'Repair Process' lays out a clear ordered sequence (quote → name cause → remove conflicts → one repair variable → conservative retry + optional variant), the 'Escalation Rules' supply a feedback loop for repeated errors, and the 'Output Contract' defines the deliverable checklist.	3 / 3
Progressive Disclosure	The body correctly defers detail to one-level-deep references signaled as [ref:...] tokens (model-mechanics, failure-atlas, delivery-qc, etc.), but no references/, scripts/, or assets/ bundle files exist, so every reference is a dangling pointer to a non-existent file and navigation is broken.	2 / 3
	Total	11 / 12 Passed

Description

90%

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

A well-targeted, third-person description with an explicit use-when trigger and strong natural-language symptom terms; its only weakness is that the stated action vocabulary is thinner than what the skill body actually delivers.

Suggestions

Broaden the stated actions beyond 'root-cause diagnosis' to include the repair and retry-generation steps the skill performs (e.g., '...diagnose the failure mode, repair the prompt, and generate a conservative retry.').

Dimension	Reasoning	Score
Specificity	It names the domain ('Seedance 2.0 output') and one concrete action ('needs root-cause diagnosis'), but does not enumerate the fuller set of actions the body performs (repair, conservative retry generation, escalation), so it stops short of the 'multiple specific concrete actions' anchor.	2 / 3
Completeness	It answers both what ('needs root-cause diagnosis') and when via an explicit 'This skill should be used when...' trigger clause, satisfying the explicit-trigger requirement that would otherwise cap this dimension at 2.	3 / 3
Trigger Term Quality	It lists a broad band of natural failure terms a user would actually say ('blurry, jittery, off-prompt, morphing, blocked, visually generic, unstable, desynced, inconsistent') plus the product name, giving strong coverage of natural trigger vocabulary.	3 / 3
Distinctiveness Conflict Risk	The narrow niche ('Seedance 2.0 output' failure modes) and highly specific symptom triggers make it unlikely to fire for an unrelated skill.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: Emily2040/seedance-2.0
Path: skills/seedance-troubleshoot/SKILL.md
Commit: 6c51262

Reviewed: 1 day 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.