Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add cclank/lanshu-awesome-ai-video-kit --skill happyhorse-promptergit clone --depth 1 https://github.com/cclank/lanshu-awesome-ai-video-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cclank/lanshu-awesome-ai-video-kit/happyhorse-prompter)<a href="https://agentmods.dev/skills/cclank/lanshu-awesome-ai-video-kit/happyhorse-prompter"><img src="https://agentmods.dev/badge/skills/cclank/lanshu-awesome-ai-video-kit/happyhorse-prompter/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cclank/lanshu-awesome-ai-video-kit/happyhorse-prompter"><img src="https://agentmods.dev/badge/skills/cclank/lanshu-awesome-ai-video-kit/happyhorse-prompter.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00116 | $0.02183 |
| Opus 5 | $0.00058 | $0.01092 |
| Sonnet 5 | $0.00023 | $0.00437 |
| Haiku 4.5 | $0.00012 | $0.00218 |
Grade A, and why
happyhorse-prompter scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
happyhorse-prompter
HappyHorse 1.0（阿里巴巴 Kling 团队）的核心差异：原生音视频联合生成 + 严格按字数执行。这个 skill 帮你产出符合它脾气的紧凑提示词。
何时不用此 skill
- 需要 10-15 秒长视频 / 多镜头切换 → 用 seedance-prompter 或 seedance-storyboard
- 复杂叙事 / 多主体 → 用 Seedance
- 已有提示词调不出来 → 还在 HappyHorse 范围内的话可以让我帮看，否则用 seedance-debugger
五大黄金规则
规则 1：主体先行，动作其后
- ❌
Walking through a forest, a woman... - ✅
A woman in hiking gear walking through a forest...
规则 2：明确指定镜头技术
HappyHorse 对"特写""广角"的理解非常精确，不要留给模型猜测。
规则 3：提示词长度控制在 30-55 词
- 过短（<30 词） → 输出模糊
- 过长（>60 词） → 模型忽略后半段
写完数一下英文单词数。
规则 4：物理效果加运动描述词
头发、水、布料、烟雾、火焰：加 slow motion、motion blur 或 fluid dynamics
规则 5：明确指定音频内容
音频激活路径（这是 HappyHorse 区别于 Seedance 的核心能力）：
with rain on leaves audible→ 雨打树叶声with engine roar audible→ 引擎轰鸣声speaking English at a natural pace→ 自然英语对话speaking Korean at a measured pace→ 韩语对话with ambient coffee shop chatter audible→ 咖啡馆环境音with the pour sound audible→ 倒水/液体声with faint crackling sound audible→ 火苗噼啪声with hoofbeats audible→ 马蹄声with drone motor whine audible→ 无人机马达声complete silence/near silence with faint wind audible→ 纯静或环境底噪
不写音频提示 = 模型可能不生成音频或乱生成。主动写。
工作流程
步骤 1：判断要不要时序节拍
如果用户的需求包含时间结构（如"先静止 2 秒，然后画面渐显"、"前 3 秒做 A，后 5 秒做 B"），用 CrePal 时序节拍写法：
8s duration. First 2s: black. Slow fade reveals: [画面 A]...
否则直接写 30-55 词紧凑版。
步骤 2：按公式拼接
[主体（明确特征）] [动作（具体）] [场景] [镜头大小+运动] [光影] [音频路径] [质量/风格].
英文写。
步骤 3：数词
- 复制到一个字数统计工具或心算 → 严格 30-55 词
- 超过 55 词：删形容词（不删核心要素），保留主体/镜头/音频路径
- 少于 30 词：补充镜头大小或音频细节
步骤 4：输出格式
## 生成的提示词
\`\`\`
[完整提示词]
\`\`\`
**词数**：N · **时长建议**：N-Ns · **比例**：[X:Y]
**音频路径**：[突出说明用了什么音频提示]
**可调点**：[换音频/换镜头/换光影都可以怎么改]
⚠️ 时长写法硬规则(每次必查)
时长信息有 2 个去处,二选一,不能同时,也不能放错位置:
| 用户场景 | 时长去处 | 写法 |
|---|---|---|
| 用户只说"做个 X 视频"(无明确时序) | 只放在外部元数据 | prompt 内不写时长;**时长建议**:5-7s 放在末尾元数据行 |
| 用户明确说"前 N 秒...然后..."(有时序节拍) | 放 prompt 最前面作时序锚点 | Ns duration, first Ns: black. [content] 这种结构 |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 175 lines · 116 tokens per session scan A 1f4b67f55119
happyhorse-prompter is a skill published in the GitHub repository cclank/lanshu-awesome-ai-video-kit (392 stars, last pushed 3mo ago), licensed MIT. It adds 116 tokens to every session and 2,183 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
cost-aware-llm-pipeline
A planning guide for choosing language models and managing the amount of conversation context used by an AI coding workflow. It groups tasks by complexity and gives rules for avoiding context overflow during long sessions.
shortfilm-prompt
Generate cinematic AI shortfilm prompts (works with Seedance 2.0, Xiaoyunque, Sora, Kling, Jimeng, Veo) using the 5-stage structure from Mx-Shell's Zombie Scavenger. Trigger when the user wants transformation sequences, multi-shot narrative shorts, weapon-charge/combat segments, emotional family/pet/farewell…
universal-video-prompt-skill
Write one model-agnostic video prompt spec, then compile it to whichever video model you can actually call. Use for cross-model prompt work, model comparison matrices, reusing one brief across providers, or when the target model is not yet available and the work must proceed on another one.
optimize-prompts
Use this to improve a prompt systematically instead of hand-tweaking it by feel. Trigger on "optimize my prompt", "make this prompt better", "the prompt isn't working well", "auto-tune my prompt", "few-shot example selection", or when prompt quality has plateaued. Optimize against an eval set with a method, and let…
agent-orchestration-improve-agent
Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.
prompt-engineer
Expert-level Prompt Engineer skill. Transforms AI into a specialist who designs, evaluates, and optimizes prompts for LLMs, RAG pipelines, and agent workflows. Covers prompt patterns (zero-shot, few-shot, CoT, ReAct, Tree-of-Thought), RAG context injection and chunking strategies, agent tool-calling and multi-agent…