CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/seedance-troubleshoot/SKILL.md
SKILL.md
Quality
Evals
Security

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.

A lean, well-structured troubleshooting skill whose diagnostic table and retry template make it genuinely actionable, with a coherent repair sequence and escalation loop. Its main weakness is progressive disclosure: the body routes to roughly a dozen reference files that are not present in the bundle and are addressed via '../..' paths outside the skill directory.

Suggestions

Ship the referenced files in a skill-local references/ directory and use relative links from SKILL.md (e.g. references/model-mechanics.md) instead of '../../references/...', or inline the small ones and cut the rest.

Add a worked example of the Conservative Retry Pattern showing one filled-in prompt for a common failure (e.g. identity drift in I2V).

Trim the 'Intent' section to one sentence and fold the tone guidance into the Output Contract to save tokens.

DimensionReasoningScore

Conciseness

The body is dense and nearly free of concept explanation Claude already knows; the 20-row diagnostic table delivers high information per token. The 'Intent' section ('A failed generation feels personal... The soul of this skill is rescue without blame...') and some abstract framing ('Remove conflicts rather than adding complexity') are minor over-explanation that could be trimmed, placing it one notch below 'every token earns its place'.

4 / 5

Actionability

The diagnostic tree (symptom → likely cause → first repair) and the fill-in 'Conservative Retry Pattern' template are concrete, executable guidance with minor gaps, matching the 'mostly executable with minor gaps' anchor. It is not a 5 because there is no worked example (e.g. a filled retry prompt) and some repairs stay generic ('Add preservation constraints') without showing what a constraint looks like.

4 / 5

Workflow Clarity

The Repair Process section gives a clear ordered sequence (quote failing phrase → state hypothesis with uncertainty → remove conflicts → one primary repair → one conservative retry), with escalation rules forming an error-recovery loop. This matches 'clear sequence with most checkpoints present; minor validation gaps' — the sequence is spread across several sections and there is no explicit validation checkpoint confirming the diagnosis before retry.

4 / 5

Progressive Disclosure

References are clearly signaled with conditional load instructions ('load [model-mechanics] when the diagnostic tree has no row'), which is good practice, but the body points to ~11 files at '../../references/*.md' and no references/ directory (or any bundle files) exist in the skill directory — the links are dangling and the paths point two levels outside the skill rather than to a skill-local bundle. Per the judging guideline to score against the actual bundle structure, the structure cannot be verified and navigation is not reliable, fitting the mid anchor rather than 'most content appropriately placed' at 4.

3 / 5

Total

15

/

20

Passed

Description

81%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.

A strong, trigger-rich description with an explicit use-when clause and a distinct Seedance 2.0 failure niche. Its main weakness is that the capability statement is reduced to a single phrase ('root-cause diagnosis'), leaving the what-it-does side implicit.

Suggestions

Add 1-2 concrete capability verbs to the 'what' side, e.g. 'Diagnoses the failed criterion, proposes a plausible root cause, and produces a conservative retry prompt.'

Optionally trim redundant symptom adjectives (e.g. 'unstable'/'inconsistent' overlap) to free tokens for the capability statement.

DimensionReasoningScore

Specificity

The description names the domain ('Seedance 2.0 output') and one concrete action ('needs root-cause diagnosis'), which matches the anchor for naming a domain with 1-2 concrete actions but lacking comprehensive coverage. It does not enumerate the actual repair actions the skill performs (diagnostic classification, repair, conservative retry prompt), so it is not at the 'several specific actions' level.

3 / 5

Completeness

The 'when' is maximally explicit ('This skill should be used when a Seedance 2.0 output is... or otherwise fails'), and the 'what' is present but thin ('needs root-cause diagnosis'). Both boxes are checked, matching the anchor where both are present but one could be more specific; it falls short of 5 because the 'what' does not concretely state what the skill does about the failure.

4 / 5

Trigger Term Quality

It lists many natural symptom terms a user would actually say — 'blurry, jittery, off-prompt, morphing, blocked, visually generic, unstable, desynced, inconsistent' — plus the catch-all 'otherwise fails'. This is comprehensive synonym coverage of natural failure language, matching the top anchor rather than merely 'good coverage with a few missing'.

5 / 5

Distinctiveness Conflict Risk

The description is tightly scoped to Seedance 2.0 video-generation failures with distinct trigger symptoms, creating a clear niche with minimal overlap risk against generic troubleshooting or prompt-writing skills.

5 / 5

Total

17

/

20

Passed

Validation

81%

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

CriteriaDescriptionResult

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

relative_links

Relative link issues: 11 suspicious

Warning

Total

13

/

16

Passed

Repository

Emily2040/seedance-2.0

Path

skills/seedance-troubleshoot/SKILL.md

Reviewed

Table of Contents

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.