Kling Prompter Agent Skill
生成符合 Kling 3.0（可灵 3.0，快手）规则的视频提示词。三种写法自适应：4 部分基础公式（短视频）/ 5 层进阶公式（剧情+音频）/ 图生视频专用（只描述运动）。Kling 是 2026 年中文理解最强、原生音画同步、最长 2 分钟、支持角色定向发声、Motion Brush 的电影级模型。用于"用 Kling 生成视频"、"可灵 AI 提示词"、"中文视频生成"、"带原生音频的剧情视频"、"图生视频"、"角色对话视频"等触发场景。
Install
one command, takes just this skill from the repositorynpx skills add https://github.com/cclank/lanshu-awesome-ai-video-kit --skill kling-prompterThe instruction itself
15 sections, as written by the authorkling-prompter
Kling 3.0（可灵 3.0，快手）的三套主流写法：4 部分基础 / 5 层进阶 / 图生视频。
何时不用此 skill
- 复杂多镜头分镜叙事（>15s） → 用 seedance-storyboard
- 5/10/15 秒紧凑短片，对原生音频要求不高 → 用 happyhorse-prompter
- 已有 Seedance 提示词需修问题 → 用 seedance-debugger
三套写法决策树
你的需求是？
├── 5 秒短视频、单一场景 → 写法 1（4 部分基础公式）
├── 带剧情、带音频、多角色 → 写法 2（5 层进阶公式）★ Kling 强项
└── 已有参考图，只需动起来 → 写法 3（图生视频专用）★ Kling 强项
工作流程
步骤 1：判断写法
| 用户的需求关键词 | 写法 |
|---|---|
| 短视频、TikTok、抖音、循环、爆款、变身 | 写法 1（4 部分） |
| 对话、剧情、配音、CTA、广告、多角色 | 写法 2（5 层） |
| 这张图、参考这张照片、让它动起来 | 写法 3（图生视频） |
步骤 2：澄清缺失要素（必要时）
如果信息不够，问 2-3 个关键问题（不是全部）：
| 场景 | 必问的 2-3 项 |
|---|---|
| 短视频 | 主体（具体外观）、动作（动词+终点）、风格（电影感/动漫/赛博朋克） |
| 5 层剧情 | 角色（年龄+服装）、对话内容、要不要原生音频（音乐/音效/voiceover） |
| 图生视频 | 主体在图中做什么变化（不要再描述图本身）、镜头怎么动 |
步骤 3：按选定写法填充
写法 1：4 部分基础公式
[Subject - 具体视觉细节] [Action - 精确运动+终点] [Context - 3-5 环境元素] [Style - 镜头+灯光+氛围].
详细参考 methodology/09-kling-公式.md。
写法 2：5 层进阶公式（Kling 的看家本领）
强烈推荐用结构化格式，模型按层解析：
Scene: [地点 + 时间 + 氛围]
Characters: [每个角色的具体描述，age + 关键服装/外貌]
Action: [按时间顺序的步骤，体现因果与递进]
Camera: [镜头类型 → 运动 → 切换]
Audio & Style: [对话/音效/音乐 + 视觉风格]
> ⚠️ Audio & Style 这层是 Kling 的差异化卖点，一定要写！Seedance / HappyHorse 都做不到角色定向发声。
写法 3：图生视频专用
铁律：只描述运动，不重复图中已有的视觉信息。
❌ 不要写：A red sports car with chrome wheels on a wet road...（图已有）
✅ 直接写：The headlights blaze on and the engine roars to life. The rear tires spin...
步骤 4：6 条 Kling 写作守则自查
- [ ] 用 active verbs 替代笼统词（rockets / slams / erupts）
- [ ] 描述 时序流（开头 → 中间 → 结尾）
- [ ] 用 具体光源 代替 "dramatic lighting"
- [ ] 加 触感细节（grain / reflections / sweat / steam）
- [ ] 加 显式 endpoint（防 99% 卡死）：
then settles back / fades out - [ ] 幻想场景用 物理交互锚定（stone crumbles / torches flicker）
步骤 5：决定语言（中文还是英文）
| 场景 | 推荐语言 |
|---|---|
| 中式美学、武侠、汉服、中文台词 | 中文（Kling 中文理解最强） |
| 国际化、Hollywood 风、产品广告 | 英文 |
| 二者混合 | 选主导风格，不混用 |
步骤 6：加负向提示
Negative: cartoonish, 3D render, smooth plastic skin, floating limbs,
sliding feet, text morphing, no watermark, no logo.
需要真人感时加 smiling 到 negative（让表情自然）。
步骤 7：输出格式
## 生成的 Kling 提示词（写法 N）
\`\`\`
[完整提示词]
\`\`\`
**推荐参数**：宽高比 [X:Y] · 时长 [N 秒] · 模型 [Kling 3.0 / Kling O1]
**关键技巧**：[1-2 句说明这条提示词的设计思路]
**可调点**：[告诉用户可以怎么改]
实战示例
例 1：用户要"短视频，女孩变成动漫角色"
判断：变身爆款 → 写法 1（4 部分基础）
输出：
A young woman in casual streetwear standing center frame, her body slowly engulfed by glowing particles as she transforms into a vibrant anime version of herself with stylized hair and large expressive eyes, then settles into her new form. TikTok-style vertical framing, neon pink and cyan rim lighting, smooth transition with sparkle bursts, high-energy K-pop MV aesthetic.
Negative: 3D render, plastic skin, floating limbs, text morphing, watermark.
参数：9:16 · 5s · Kling 3.0
关键技巧：4 要素齐全 — Subject（具体服装）+ Action（带终点：then settles）+ Context（3 个：neon rim light/sparkle/vertical）+ Style（K-pop MV aesthetic）。
可调点：把 K-pop MV 换成 Cyberpunk / Studio Ghibli 改风格；把 glowing particles 换成 glitch effects 改变身机制。
例 2：用户要"两个老朋友 10 年后重逢，要带配音"
判断：带剧情 + 带音频 → 写法 2（5 层进阶）
输出：
Scene: A park bench at golden hour in late autumn, fall leaves drifting through the air, soft amber sunlight slanting through bare branches.
Characters: Two reunited friends with a 10-year gap, both early 30s. Character A wears a wool coat and scarf, slightly nervous. Character B wears a denim jacket, more outwardly emotional.
Action: Character A sits down beside Character B, hesitates, then says "It's been a long time." Character B turns slowly, eyes widening with recognition, replies "Too long. I almost didn't recognize you." They both pause, then share a quiet laugh that breaks the tension.
Camera: Slow push-in on Character A as he speaks, then over-the-shoulder cut to Character B's reaction, ending on a held wide two-shot of both on the bench.
Audio & Style: Soft solo piano underscore building gently, ambient park sounds (distant kids playing, leaves rustling), warm cinematic color grade with teal-and-amber accents, naturalistic dialogue delivery.
Negative: smiling broadly, cartoonish, 3D render, plastic skin, sliding feet, watermark.
参数：16:9 · 12s · Kling 3.0（推荐）
关键技巧：5 层全开。Characters 层给两个角色各自的「锚点服装」防漂移；Action 层用因果链（坐下 → 犹豫 → 开口 → 反应 → 共同笑）让叙事自然；Audio & Style 层既给音乐又给环境音，触发 Kling 原生音画同步。
可调点：换 golden hour → blue hour 改基调；换 piano underscore → silence + only ambient 让情感更克制。
例 3：用户上传了一张厨房照片，想"让它动起来，厨师颠炒"
判断：图生视频 → 写法 3
输出：
The chef flips the wok hard, launching noodles into the air as a fireball erupts upward. He catches them clean, tilts the wok and slides the dish onto a plate with his spatula, then pushes the plate across the worn counter toward the camera with a grin. Steam billows upward through the lantern light.
Negative: floating limbs, plastic skin, sliding feet, watermark.
参数：16:9 · 8s · Kling 3.0
关键技巧：完全没重描述图中的厨房布局/厨师外貌（图自己有），只写运动序列（flip → fire → catch → slide → push）+ 物理细节（steam billows）。明火创造动态光照是 Kling 的强项。
可调点：删 fireball erupts 改成温和炒制；把 with a grin 删掉让表情更专注。
参考资源
- Kling 完整方法论：methodology/09-kling-公式.md
- 36 条实测提示词：prompts/kling/README.md
- 运镜词典（通用）：methodology/05-运镜词典.md
- 官方平台：klingai.com
Other skills for the same job
different authors, same section of the catalogue
agent-im 会话技能 - 通过 liblib.tv 的 AI 能力生成和编辑图片/视频。覆盖场景包括：生成（文生图、文生视频、图生视频、做动画、画一个xxx、来段xxx）、编辑修改（把xxx换成yyy、去掉xxx、加上xxx、改成xxx、调整xxx、局部修改、改镜头）、风格转换（风格迁移、转绘、换风格）、视频续写延长、复刻视频/TVC/宣传片、短剧/短漫剧生成、音乐MV生成、产品广告/展示片制作、分镜/故事板设计、教育视频/短视频制作。当用户提到 liblib、libtv、上传参考图/视频、查看生成进度时也应触发。关键判断：只要用户的请求涉及 AI 图片或视频的创作、生成、编辑、修改，无论措辞如何（如"画只猫"、"做个海报"、"把纸船换成爱心"、"这个视频帮我改一下"、"帮我复刻这段视频"、"用这首歌做个MV"、"一句话生成短剧"），都必须触发此技能。
This skill should be used when the user asks to "generate video prompts", "create Seedance prompts", "write video descriptions", mentions "Seedance", "seedance", "即梦", "即梦平台", "视频提示词", "视频生成", "AI视频", "短剧", "广告视频", "视频延长", or discusses video prompt engineering, AI video generation, or Seedance 2.0 workflows.
Best practices and techniques for writing effective AI video generation prompts. Covers: Veo, Seedance, Wan, Grok, Kling, Runway, Pika, Sora prompting strategies. Learn: shot types, camera movements, lighting, pacing, style keywords, negative prompts. Use for: improving video quality, getting consistent results, professional video prompts. Triggers: video prompt, how to prompt video, veo prompts, video generation tips, better ai video, video prompt engineering, video prompt guide, video prompt template, ai video tips, video prompt best practices, video prompt examples, cinematography prompts
This skill is a practical, 'use-it-while-debugging' reference for getting a LiveKit + Letta voice agent working reliably.
Download screenshot baselines from the latest CI run and commit them. Use when asked to update, accept, or refresh component screenshot baselines from CI, or after the screenshot-test GitHub Action reports differences. This skill should be run as a subagent.
| Turn vague taste, screenshots, URLs, product notes, or "make it feel like this" references into a grounded DESIGN.md plus an implementation handoff. Use it before prototypes, decks, redesigns, or image remix work when the user needs a reusable visual direction rather than a one-off prompt.
>- Upload local assets (images, mockups, extracted HTML, design markdown) to a Stitch project. ALWAYS use this skill when you need to upload visual assets, HTML pages, or design docs to Stitch, particularly when direct MCP tool calls fail or truncate due to base64 token limits.
This skill helps users automatically extract channel-level and video detail data from a specific YouTube channel via BrowserAct API. Agent should proactively apply this skill when users express needs like extracting channel video data, getting latest or popular videos from a YouTube channel, tracking competitor channel content, extracting video metrics such as views likes comments, retrieving subscriber count and channel info, monitoring posting cadence of a YouTube channel, gathering video data for content strategy analysis, getting earliest videos of a YouTube creator, analyzing engagement signals across a full channel, and downloading structured YouTube video details without manual scraping.
How to use it
Take cclank/kling-prompter from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.