prompt-translator
Prompt Translator converts video prompts across models by matching your input to one of 10 reference scenarios, then applying the target model's format rules from a lookup table of 110 cross-model examples. It preserves all semantic content while restructuring labels and phrasing to fit each model's official formula.
Prompt Translator converts your video prompt from one AI model's format to another's best practices using reference examples.
AI-generated summary based on this skill's SKILL.md
Install
cclank/lanshu-awesome-ai-video-kit/prompt-translator · repository language: HTML
git clone https://github.com/cclank/lanshu-awesome-ai-video-kit
cp -r lanshu-awesome-ai-video-kit/skills/prompt-translator ~/.claude/skills/prompt-translatorFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
How do I translate AI video prompt between models using Prompt Translator?
Prompt Translator converts your video prompt by first matching it to one of 10 reference scenarios, then applying your target model's format rules from a lookup table of 110 cross-model examples. The tool preserves all semantic content while restructuring labels and phrasing to fit each model's official formula, so your creative intent stays intact across platforms.
Can Prompt Translator convert a Sora prompt to Kling format?
Yes. Prompt Translator handles conversion from Sora to Kling and between most major video generation models. It restructures your Sora prompt's labels and phrasing to match Kling's best practices while keeping all semantic meaning. The tool draws from 110 cross-model examples to ensure accurate format translation.
What does Prompt Translator do with my prompt's semantic content?
Prompt Translator preserves all semantic content during conversion. When you adapt a prompt from one video model to another, the tool only changes structural tags and field labels to match the target model's format—your creative direction, visual descriptions, and technical specifications remain unchanged.
Which video models does Prompt Translator support?
Prompt Translator works across major video AI platforms including Sora, Kling, Veo, Pika, Wan, and Seedance. The tool uses 110 cross-model examples and matches your input to one of 10 reference scenarios, then applies the correct format rules for your chosen target model.
Can I use Prompt Translator for A/B testing across models?
Yes. Prompt Translator enables cross-model A/B testing by translating your prompt to multiple target formats simultaneously. This lets you run the same creative concept on different video generation platforms while maintaining semantic consistency, helping you compare model outputs fairly.
Does Prompt Translator support language optimization like English to Chinese?
Prompt Translator includes language and dialect optimization. It can adapt prompts between English and Chinese Kling, and similar language variants across supported models. The tool restructures both format and phrasing to match each model's regional best practices while preserving your original intent.
SKILL.md
rendered from the published skill — quoted content, verbatim
prompt-translator
跨模型提示词转换器。关键差异:不是凭 AI 直觉重写,而是查 110 条对照基准做 in-context learning。
何时不用此 skill
- 用户从零开始写一条新提示词(不是转换) → 用
seedance-prompter/kling-prompter/happyhorse-prompter或model-selector - 用户问"用哪个模型好" → 用
model-selector - 已有提示词出问题但不换模型 → 用
seedance-debugger
核心数据资产
prompts/data/cross-model-matrix.json — 这就是 translator 的"训练数据":
- 10 个核心场景:产品广告 / 情感重逢 / 滑板动作 / 图生视频 / 多人会议 / 恐怖悬疑 / 自然延时 / 抽象艺术 / 武侠决斗 / 萌宠爆款
- 每个场景 × 11 模型 = 110 条对照 prompt,每条严格遵循对应模型的官方公式
- 这构成"同一场景在 11 模型上的最佳写法对照",就是 translator 的查找表
工作流程
步骤 1:接收输入
最少需要: - 源模型 (如 Sora 2 / Kling 3.0 / Wan 2.7) - 源 prompt (用户的现有提示词) - 目标模型 (用户想转到哪个)
可选: - 转换偏好(更简洁 / 更详细 / 保留中文)
步骤 2:分析源 prompt
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 1 file
skills/prompt-translator/SKILL.md