CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

muapi-seedance-2

Expert Cinema Director skill for Seedance 2.0 (ByteDance) — high-fidelity video generation across Chinese, Global, and VIP tiers. Supports text-to-video, image-to-video, first-last-frame, omni reference, character training, omni-reference training, video editing, and watermark removal.

Quality

71%

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 ./library/motion/seedance-2/SKILL.md

The canonical home for this skill is muapi-seedance-2 in SamurAIGPT/Generative-Media-Skills

SKILL.md
Quality
Evals
Security

Quality

Content

57%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 highly actionable with complete executable examples across every mode, but it is verbose (padding with basic cinematography definitions and repeated warnings) and inlines large reference tables that would be better split into separate files. Workflow guidance lacks explicit validation/feedback loops for its destructive operations.

Suggestions

Move the Camera Language Reference, Shot Sizes, and Style & Quality Modifiers tables into a separate references/ file (e.g. camera-language.md) and link to it from the body, cutting substantial inline tokens.

Deduplicate the 'audio is mandatory' and 'single-beat' guidance — state each once in the Prompt Optimization Protocol and reference it rather than repeating across Common Mistakes and examples.

Add an explicit validation/feedback loop for video-edit and watermark-remove modes (e.g. check the returned video, retry with adjusted prompt if artifacts or leftover watermarks appear).

DimensionReasoningScore

Conciseness

Noticeably verbose: a basic cinematography glossary ('Pan left/right = Camera rotates horizontally', 'Close-up = Face fills frame') explains concepts Claude already knows, and key warnings are repeated multiple times (audio mandatory appears ~4 times, single-beat rule ~3 times).

2 / 5

Actionability

Fully executable copy-paste bash commands with complete flags cover all nine modes and common cases, plus concrete subject-prompt examples for each pattern.

5 / 5

Workflow Clarity

A numbered Prompt Optimization Protocol and an async submit/check-status pattern give a sequence, but there are no validate→fix→retry feedback loops for destructive or batch operations (video-edit, watermark-remove), capping workflow clarity at 3.

3 / 5

Progressive Disclosure

Section structure is clear and the single bundle script (scripts/generate-seedance.sh) is well signaled, but substantial reference material (camera glossary, shot sizes, style modifiers, prompt templates) is inlined rather than split into reference files; only an executable bundle exists, no reference docs.

3 / 5

Total

13

/

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.

The description is specific and highly distinct, enumerating eight concrete capabilities with a clear product niche. Its main weakness is the absence of an explicit 'Use when...' trigger clause, which caps completeness.

Suggestions

Append an explicit trigger clause, e.g. 'Use when generating or editing video with Seedance 2.0, or when the user mentions text-to-video, image-to-video, first-last-frame, or video editing.'

Add a couple of natural synonyms (e.g. 'AI video', 'video generation', 'animate images') alongside the mode names to broaden trigger coverage.

DimensionReasoningScore

Specificity

Lists multiple concrete actions comprehensively — 'text-to-video, image-to-video, first-last-frame, omni reference, character training, omni-reference training, video editing, and watermark removal' — covering the full capability surface.

5 / 5

Completeness

The 'what' is clear and detailed, but there is no explicit 'Use when...' trigger clause, so per the rubric cap completeness cannot exceed 3.

3 / 5

Trigger Term Quality

Good natural keywords ('video generation', 'text-to-video', 'image-to-video', 'video editing', 'watermark removal') but leans on model-specific jargon (Seedance 2.0, ByteDance) and omits common synonyms or file extensions.

4 / 5

Distinctiveness Conflict Risk

Names a clear niche (Seedance 2.0 / ByteDance with Chinese/Global/VIP tiers) with distinct triggers and minimal overlap risk with other skills.

5 / 5

Total

17

/

20

Passed

Validation

87%

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

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (633 lines); consider splitting into references/ and linking

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

SamurAIGPT/Generative-Media-Skills

Path

library/motion/seedance-2/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.