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

76%

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 ./.opencode/skills/muapi-seedance-2/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.

An extremely actionable, well-organized reference whose commands and prompt patterns are immediately usable. It is weakened by monolithic inlining (with the referenced script absent from the bundle), duplicated audio/single-beat reminders, and a camera-basics glossary that assumes Claude lacks common cinematography knowledge.

Suggestions

Move the camera-language glossary, prompt template library, and endpoint/parameter tables into reference files (e.g. references/camera-language.md, references/templates.md, references/endpoints.md) and link them from a lean SKILL.md overview.

Ensure 'scripts/generate-seedance.sh' actually exists in the bundle (and fix or explain the relative '../../../../../core/media/generate-video.sh' path in the async pattern), since every example depends on it.

State the audio-mandatory and single-beat rules once each (removing the duplicates in the Optimization Protocol and Common Mistakes sections) and trim basic shot-size/camera-movement definitions Claude already knows.

DimensionReasoningScore

Conciseness

Mostly efficient model-specific reference (tiers, limits, @-system, endpoints), but it defines camera basics Claude already knows ('Pan left/right — Camera rotates horizontally', 'Close-up — Face fills frame') and repeats the audio-mandatory rule three times and the single-beat rule twice.

3 / 5

Actionability

Every one of the nine modes ships a copy-paste-ready 'bash scripts/generate-seedance.sh' command with real flags, plus concrete worked Director Brief examples per capability — fully executable guidance covering the common cases.

5 / 5

Workflow Clarity

Each mode has a clear command sequence, the async submit→poll pattern is shown, and the omni-train → '@omni-character:<id>' handoff is explained; minor gaps include no explicit checkpoint for confirming training completed and only implicit status/result verification.

4 / 5

Progressive Disclosure

Well-sectioned headers, but it is a ~630-line monolith: the camera-language reference, prompt templates, and endpoint tables belong in separate files, and every example invokes 'scripts/generate-seedance.sh' which does not exist in the bundle (no scripts/, references/, or assets/ directories).

3 / 5

Total

15

/

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 highly specific, well-scoped description that comprehensively enumerates capabilities and is firmly third-person. Its one structural weakness is the absence of any 'when to use this' trigger clause, which limits discoverability and caps completeness.

Suggestions

Append an explicit trigger clause, e.g. 'Use when the user wants to generate, animate, extend, or edit AI video with Seedance 2.0, or mentions video generation, image-to-video, or watermark removal.'

Add natural user phrasings and synonyms ('make a video', 'animate this image', 'AI video clip') to broaden trigger-term coverage.

Clarify what 'Chinese, Global, and VIP tiers' means for a user choosing the skill, or drop the tier mention from the description since it is explained in the body.

DimensionReasoningScore

Specificity

Lists nine concrete capabilities ('text-to-video, image-to-video, first-last-frame, omni reference, character training, omni-reference training, video editing, and watermark removal') plus a tier breakdown — comprehensive, specific action coverage with no gaps.

5 / 5

Completeness

The 'what' is explicit and detailed, but there is no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Good natural keyword coverage ('video generation', 'video editing', 'watermark removal', 'Seedance 2.0') but misses common user phrasings like 'generate a video', 'animate an image', or format synonyms, so it falls short of the comprehensive anchor.

4 / 5

Distinctiveness Conflict Risk

Scoped to a named model and version ('Seedance 2.0 (ByteDance)') with distinctive capabilities (omni reference, character training, watermark removal) — a clear niche with minimal overlap risk against generic video 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

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

referenced_paths_exist

Referenced path issues: 26 missing

Warning

Total

13

/

16

Passed

Repository

SamurAIGPT/Generative-Media-Skills

Path

.opencode/skills/muapi-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.