lanshu-awesome-ai-video-kit skill
Seven Claude Code skills and a prompt library for the Seedance, Kling and HappyHorse video models: prompts, storyboards, debugging
- 437 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- The skills are plain SKILL.md instruction files with no scripts, and the viewer files run only locally
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/lanshu-awesome-ai-video-kitDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skills are plain SKILL.md instruction files with no scripts, and the viewer files run only locally
This is third-party code. Review the repository files before installing.
What it does
lanshu-awesome-ai-video-kit is a collection for writing prompts for video models, which the author put together on corporate video projects. The repository holds a library of ready prompts by model and scenario, 21 methodology write-ups (formulas, storyboarding, camera language, constraint words, common pitfalls) and seven Claude Code skills. The seedance-prompter skill builds a structured Seedance prompt, seedance-storyboard splits a plot into timed shots, seedance-debugger finds why a prompt failed across 12 categories, and happyhorse-prompter and kling-prompter write prompts for HappyHorse and Kling. The model-selector skill asks a few questions and recommends one to three models, and prompt-translator rewrites a prompt from one model's formula into another's. Skills install by copying folders into ~/.claude/skills/, after which Claude Code picks the right one from a phrase like «make a product ad video». The prompt library is easiest to browse in a single-page viewer: run `python3 -m http.server 8000` in the repository folder and open /tools/prompt-browser/.
Who it is for. For creators, marketers and developers who make clips in Seedance, Kling or HappyHorse and want Claude Code to write and debug prompts.
Good fit when
- You need a product ad clip and want the agent to build a structured Seedance prompt
- A plot has to be split into 3–5 timed shots for a video model
- A prompt does not give the expected result and you need a 12-category diagnosis with a fixed version
- You must choose between Sora, Kling or another model given length, audio, language and budget
- A finished prompt for one model needs porting to another model's formula
Not a fit when
- You need the video generation itself: the kit writes prompts and does not connect to models
- You need English-language skill instructions: the SKILL.md files are written in Chinese
- You work with a model that has no dedicated skill: prompts and write-ups exist for it, but the agent guides you only for Seedance, HappyHorse and Kling
Example request
Make a 10-second ad clip for a new coffee machine: choose a model and write a Seedance prompt with a three-shot storyboardLimitations
- The skills sit in the repository as separate folders. There is no ready install command: copy them into the skills directory with `cp -r`.
- Skill instructions and methodology write-ups are in Chinese, and the README comes in English and Chinese.
- The model table is a May 2026 snapshot refreshed monthly. Its Sora 2 row already notes that the API was shut down on 24 September 2026. A weekly GitHub Actions job watches the official endpoints and opens issues when versions change.
- The kit itself makes no video. You need access to the chosen model, and its terms and price depend on the provider.
- Model ratings in the tables, such as stars for language, audio and physics, are the author's own.
How to disable. Delete the folders you copied from ~/.claude/skills/ and restart Claude Code. The kit leaves no other traces.
FAQ
How do I install the lanshu-awesome-ai-video-kit skills in Claude Code?
Clone the repository and copy the folders you need from skills/ into ~/.claude/skills/, for example `cp -r skills/kling-prompter ~/.claude/skills/`. After a restart Claude Code picks the skill from your phrase.
Which skill do I use if I do not know which video model suits me?
The model-selector skill. It asks three or four questions, such as scene, length, audio, language, budget and local deployment, and recommends one to three models with reasoning. It triggers on questions like «Sora or Kling?».
Can I port a Sora prompt to Kling?
Yes, prompt-translator does that. It triggers on a request like «translate this prompt to Kling style» and returns a prompt in the target model's formula, a mapping table and caveats. It builds on 110 cross-model prompts for 11 commercial models.
What does seedance-debugger do?
It analyzes a prompt that does not work across 12 categories of causes and returns a fixed version. The methodology also has a write-up of 12 common pitfalls, each with symptom, root cause and fix.
How do I open the prompt library in a browser?
In the repository folder run `python3 -m http.server 8000` and open http://localhost:8000/tools/prompt-browser/. The viewer filters by scenario, model and tags, and a chosen filter can be shared as a link.
Discussion
No questions or reviews yet.
Related
SkillsA Claude Code skill that turns an idea into a five-stage prompt for Sora, Kling, Veo and Seedance, with templates for 21 genres
A set of skills for Claude Code and Codex that assemble detailed image prompts from an element library and composition schemes
A library of 280-plus skills and annotated cases for writing short-video prompts in MiniMax H3 and Seedance 2.0
A skill that writes prompts for video, image and music generation on 14 platforms and can submit them in the browser
Higgsfield Claude Skills: 19 skills for video and UGC ads
Higgsfield Claude Skills
19 Claude Code skills for Higgsfield: Seedance 2.0 prompts in 15 styles and generation automated through Playwright
Two skills for generative video and images: video writes Seedance, Kling and Veo prompts by dramaturgy rules, image picks the model and its syntax
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| lanshu-awesome-ai-video-kit (this entry) | no | 437 | |
| Seedance 2.0 Prompt Writing Skills | no | 4,376 | a single Seedance 2.0 skill with @ references to material and templates, while this kit also covers HappyHorse and Kling, model choice and prompt porting |
| ai-shortfilm-prompts | no | 499 | turns an idea into a five-stage prompt from 21 genre templates, while this kit offers a prompt library and debugging for specific models |
README in short
The page explains how prompting a video model resembles directing a shot and gathers the vendors' official guides. Then come a model table, a quick start, the list of methodology documents, a skills table, web tools and the directory structure.
README badge
Original SKILL.md
--- name: model-selector description: 根据用户的视频需求(场景、时长、音频、语言、平台限制、预算、是否需要本地部署/角色一致性等),从 16 个主流 AI 视频模型(12 商业 + 4 开源)中推荐最匹配的 1-3 个,并解释为什么。覆盖商业:Seedance 2.0、HappyHorse 1.0、Kling 3.0、Sora 2、Veo 3.1、**Gemini Omni**(2026-05 新)、Runway Gen-4.5/Aleph、Pika 2.5、Hailuo 02、Hunyuan Video 1.5、Wan 2.7、即梦 AI;开源:LTX-Video 0.9.7、Mochi 1、CogVideoX 5B、Higgsfield Soul。用于"用哪个 AI 视频模型好"、"Sora 还是 Kling"、"国内有什么模型"、"哪个免费"、"哪个支持中文"、"哪个能编辑已有视频"、"哪个能本地部署"、"哪个角色一致性最强"等触发场景。 --- # model-selector 15 个主流 AI 视频模型(11 商业 + 4 开源)的"购物顾问"。问用户 3-4 个关键问题,给出推荐 + 理由。 ## 何时不用此 skill - 用户已经确定要用某个模型 → 用对应的 `<model>-prompter` skill - 用户问"提示词怎么写" → 用 `seedance-prompter` / `kling-prompter` / `happyhorse-prompter` - 用户的提示词不工作 → 用 `seedance-debugger` ## 11 模型能力速查矩阵 | 模型 | 中文 | 音频 | 物理 | 编辑 | 时长 | 开源 | 国内可用 | 主战场 | |---|---|---|---|---|---|---|---|---| | **Seedance 2.0** | ★★★ | — | ★★★ | — | 15s | — | ✓ | 电影叙事 | | **HappyHorse 1.0** | ★★★ | ★★ | ★★ | — | 15s（默认 5s） | — | ✓ | ASMR / 短片 | | **Kling 3.0** | ★★★★★ | ★★★★ | ★★★★ | ✓ | 2m | — | ✓ | 中文剧情 / 图生视频 | | **Sora 2** | ★★ | ★★★★ | ★★★★★ | — | 25s (Pro) | — | △ | 电影艺术片 | | **Veo 3.1** | ★★ | ★★★★★ | ★★★ | — | 148s chained | — | △ | 多人对话 | | **Gemini Omni** ⭐NEW | ★★ | ★★★ | ★★★ | ★★★★★ | 待定 | — | △ | **迭代编辑 / 文字渲染 / 跨模态** | | **Runway Gen-4** | ★★★ | — | ★★★ | ★★★★★ | 30s | — | ✓ | **视频编辑** | | **Pika 2.5** | ★★ | ★ | ★★ | — | 25s | — | ✓ | **创意特效** | | **Hailuo 02** | ★★★ | — | ★★★★★ | — | 10s | — | ✓ | 物理动作 | | **Hunyuan 1.5** | ★★★★ | — | ★★★ | ✓ | 10s | ★★★★★ | ✓ | **开源 / LoRA** | | **Wan 2.7** | ★★★ | ★★★★★ | ★★★ | ✓ | 15s | — | ✓ | **数字人 lip-sync** | | **即梦 3.0** | ★★★★★ | ★★ | ★★★ | — | 15s | — | ✓ | **中文 + 剪映集成** | ## 工作流程 ### 步骤 1：问 3-4 个关键问题 不要全部问，挑最影响结果的 3-4 个： | 维度 | 问题 | |---|---| | 时长 | "需要多长？5 秒 / 10 秒 / 15 秒以上？" | | 音频 | "要原生音频吗？只要环境音 / 要对白 / 要 BGM / 不要" | | 语言 | "是中文场景还是英文场景？" | | 编辑性质 | "全新生成还是编辑已有视频？" | | 平台限制 | "国内可用 / 国际可用 / 本地部署？" |