CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

seedance-prompt-short

This skill should be used when the user asks for a compact Seedance 2.0 prompt, short Chinese prompt, prompt compression, 30-100 word output, or removal of unnecessary prompt language.

Quality

74%

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-prompt-short/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

65%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.

An impressively dense, non-padded body with genuinely actionable templates and a concrete preservation-order list. Its weaknesses are an implicit, non-enumerated workflow and a hard dependency on six reference files whose paths resolve outside the skill directory and are absent from the bundle.

Suggestions

Add a short numbered workflow (load Director's Read → classify brief → complete record → compress per priority → run moderation pre-screen → deliver per Output Contract) so the sequence does not have to be reconstructed from prose, and consolidate the duplicated Director's Read loading instruction from the first and third paragraphs into that single step.

Fix the reference paths: all six citations use '../../references/*.md', which escapes the skill directory, and none of the files exist in the bundle — a user invoking the skill as shipped cannot perform its mandatory first step.

Trim the 'Intent' section's aphorisms ('If the user would mourn a deleted word, it was never filler', 'Short means compressed output, not skipped judgment') to one-line principles so the tokens they occupy go to executable rules.

DimensionReasoningScore

Conciseness

The body is lean with no padding or explanation of concepts Claude already knows, and the Compact Templates table earns its tokens. Minor trimmable fat remains: the Director's Read loading instruction is stated twice (first and third paragraphs), and aphorisms like 'If the user would mourn a deleted word, it was never filler' are flourish rather than instruction.

4 / 5

Actionability

The Compact Templates table provides copy-paste-ready skeletons for T2V, I2V, V2V, Chinese, and storyboard cases, and the numbered Compression Priority list is concrete and ordered. However, key operational detail (the ten-field narrative record, shot-table cells) is deferred to reference files that do not exist in this bundle, leaving gaps.

4 / 5

Workflow Clarity

A rough sequence exists in prose order (load Director's Read → classify the brief → complete the record → compress → pre-screen → deliver) and checkpoints are mentioned ('The moderation pre-screen runs on the compressed prompt too', 'a cell it cannot fill is a question it must ask'), but no steps are enumerated and the checkpoints remain implicit rather than explicit validation gates.

3 / 5

Progressive Disclosure

References are clearly signaled inline and the body appropriately stays an overview, but all six point to '../../references/' paths that escape the skill directory, and none of the referenced files (directors-read.md, moderation-prescreen.md, shot-table.md, direct-for-the-model.md, multishot-grammar.md, multilingual-community-examples.md) exist in the bundle — the navigation is signposted but not actually loadable, and the mandatory first step cannot be completed as shipped.

3 / 5

Total

14

/

20

Passed

Description

75%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 solid trigger-oriented description: explicit 'use when' guidance with natural keywords and a clear Seedance 2.0 niche. It falls short of the top anchors because the capability is implied by triggers rather than stated, and coverage synonyms are thin.

DimensionReasoningScore

Specificity

Names the domain ('Seedance 2.0') and several concrete capabilities — 'compact Seedance 2.0 prompt', 'short Chinese prompt', '30-100 word output', 'removal of unnecessary prompt language' — but these largely restate one operation and omit coverage of storyboards, I2V/V2V, or bilingual work that the body actually handles.

4 / 5

Completeness

The 'when' is fully explicit ('This skill should be used when the user asks for...'), but the 'what' is never stated as an action — there is no third-person statement like 'Compresses Seedance 2.0 prompts to 30-100 words'; the capability must be inferred from the trigger list.

4 / 5

Trigger Term Quality

'prompt compression', 'short Chinese prompt', 'compact Seedance 2.0 prompt', and '30-100 word output' are phrases a user would naturally say. Common synonyms such as 'shorten', 'condense', or 'trim a prompt' are missing, keeping it below the comprehensive anchor.

4 / 5

Distinctiveness Conflict Risk

'Seedance 2.0' carves a clear niche with distinct triggers, but generic phrases like 'prompt compression' and 'removal of unnecessary prompt language' overlap with any general prompt-shortening skill and with the apparent parent skill (metadata.parent: seedance-20).

4 / 5

Total

16

/

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: 7 suspicious

Warning

Total

13

/

16

Passed

Repository

Emily2040/seedance-2.0

Path

skills/seedance-prompt-short/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.