Agent SkillsEmily2040/seedance-2.0 › seedance-prompt-short

seedance-prompt-short

GitHub

用于将 Seedance 2.0 提示词压缩至 30-100 字。通过加载导演阅读材料分类并保留核心叙事与连续状态,去除冗余描述,生成紧凑且符合生产要求的短提示词。

skills/seedance-prompt-short/SKILL.md Emily2040/seedance-2.0

Trigger Scenarios

请求简短的 Seedance 提示词 要求压缩提示词长度 移除多余提示语言

Install

npx skills add Emily2040/seedance-2.0 --skill seedance-prompt-short -g -y
More Options

Use without installing

npx skills use Emily2040/seedance-2.0@seedance-prompt-short

指定 Agent (Claude Code)

npx skills add Emily2040/seedance-2.0 --skill seedance-prompt-short -a claude-code -g -y

安装 repo 全部 skill

npx skills add Emily2040/seedance-2.0 --all -g -y

预览 repo 内 skill

npx skills add Emily2040/seedance-2.0 --list

SKILL.md

Frontmatter
{
    "name": "seedance-prompt-short",
    "tags": [
        "prompt-compression",
        "chinese-prompt",
        "seedance-20"
    ],
    "license": "MIT",
    "metadata": {
        "author": "Iamemily2050 (@iamemily2050)",
        "parent": "seedance-20",
        "updated": "2026-08-01",
        "version": "6.7.0",
        "openclaw": {
            "emoji": "🎬",
            "homepage": "https:\/\/github.com\/Emily2040\/seedance-2.0"
        },
        "repository": "https:\/\/github.com\/Emily2040\/seedance-2.0"
    },
    "description": "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.",
    "user-invocable": true
}

seedance-prompt-short

Before producing prompt text, a prompt-ready block, a rewrite, an example, or a compiled clip, load the Director's Read, classify the brief, and complete its canonical narrative or non-narrative record. Translate that record into visible or audible carriers and keep its internal labels out of final generation prose.

Compress Seedance prompts without losing the production signal. A short prompt still needs mode, subject, action, camera, light, sound when useful, and constraints. Remove filler before removing physical details.

Load the Director's Read before compression, even when the requested output is only one line. Narrative, story, and performance briefs complete its ten-field internal record first; non-narrative utility, product-only, abstract, VFX, and ambient briefs use its two-line utility intent and refusal without fabricated drama. Short means compressed output, not skipped judgment.

When sequence state is present, compression must preserve continuity locks, exact reference tags, actual opening state, current clip action, endpoint, completed beat exclusions, and reserved future beats. Do not compress away the words that keep a continuation from replaying completed action or leaking future action.

Intent

Compression is an act of judgment about what the user loves most. What survives the cut is the soul of their shot; everything else goes first. If the user would mourn a deleted word, it was never filler.

Compression Priority

Preserve in this order:

  1. Reference tags and their role.
  2. Subject or product identity.
  3. Action verb and visible endpoint.
  4. For narrative work, the visible suppressed behavior and non-transferable detail that carry the Director's Read.
  5. One camera move.
  6. Physical light source or atmosphere.
  7. Audio cue or silence instruction.
  8. Safety, IP, or continuity constraint.
  9. Sequence state clauses: actual opening state, continuity locks, completed beats, and reserved beats.

Delete generic adjectives, duplicate style labels, obvious background details, secondary camera moves, and secondary actions before deleting preservation constraints.

For bilingual or mixed-language compression, load multilingual-community-examples. Keep only the language mix that clarifies reference roles, dialogue, camera terms, or safe production constraints.

Compact Templates

Need Template
T2V [Subject] [action and endpoint] in [scene]. Camera: [one move]. Light/style: [physical source]. Sound: [cue]. Constraint: [risk/continuity].
I2V @Image1 preserved; only [motion/light/camera] changes. Camera: [one move]. Sound: [cue]. Constraint: [what must not change].
V2V @Video1 controls [motion/camera/timing] only; new subject [anchor]. [Action]. Do not transfer [identity/scene/logo].
Chinese @Image1为参考,严格保持[主体]不变;仅加入[动作/光线/镜头]。声音:[提示]。

Output Contract

Return one compact prompt, ideally 30-100 English words or an equivalent Chinese prompt when the user asks for Chinese or maximum compression. Translate the Director's Read into visible or audible carriers and keep its internal labels out of the prompt. Include a one-line note only if something important was removed.

Version History

  • b8cdc5b Current 2026-08-04 19:06

    新增强制加载Director's Read逻辑以规范提示词路由;修复嵌套运行时路由的可移植性问题;引入便携式导演阅读机制。

  • 7f22eac 2026-08-02 22:01

    更新版本至 v6.7.0,记录 Seedance 2.5 信息并修正前端排版与字体规范问题。

  • 6c51262 2026-07-30 20:23

Same Skill Collection

skills/seedance-antislop/SKILL.md
skills/seedance-audio/SKILL.md
skills/seedance-camera/SKILL.md
skills/seedance-characters/SKILL.md
skills/seedance-continuation/SKILL.md
skills/seedance-copyright/SKILL.md
skills/seedance-examples-ja/SKILL.md
skills/seedance-examples-ko/SKILL.md
skills/seedance-examples-zh/SKILL.md
skills/seedance-filter/SKILL.md
skills/seedance-interview-short/SKILL.md
skills/seedance-interview/SKILL.md
skills/seedance-lighting/SKILL.md
skills/seedance-motion/SKILL.md
skills/seedance-pipeline/SKILL.md
skills/seedance-prompt/SKILL.md
skills/seedance-recipes/SKILL.md
skills/seedance-sequence/SKILL.md
skills/seedance-style/SKILL.md
skills/seedance-troubleshoot/SKILL.md
skills/seedance-vfx/SKILL.md
skills/seedance-vocab-en/SKILL.md
skills/seedance-vocab-es/SKILL.md
skills/seedance-vocab-ja/SKILL.md
skills/seedance-vocab-ko/SKILL.md
skills/seedance-vocab-ru/SKILL.md
skills/seedance-vocab-zh/SKILL.md
SKILL.md

Metadata

Files
0
Version
f084a54
Hash
3225ea5b
Indexed
2026-07-30 20:23

Accueil - Wiki
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-08-07 07:18
浙ICP备14020137号-1 $Carte des visiteurs$