Realistic Video Prompting logo

Realistic Video Prompting

Community
zhouwei713
realistic-video-prompting

为AI视频生成模型(Seedance、Sora、Kling、Runway、Veo等)撰写真实感、去AI感、纪录片质感的结构化视频提示词。当用户想生成AI视频提示词、优化现有视频提示词、追求"真实感""家庭录像感""手机随拍感""DV感""监控录像感"的视频效果,或提到Seedance、文生视频、视频prompt时,务必使用此skill。即使用户只是描述了一个想拍的场景或氛围,也应使用此skill将其转化为完整提示词。

Overview

Publisherzhouwei713
Repositoryseedance-prompt
Skill namerealistic-video-prompting
Stars
330
Forks
55
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by zhouwei713 on GitHub. Read the source before you install it.

Installation

Install the Realistic Video Prompting AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/zhouwei713/seedance-prompt/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/zhouwei713/seedance-prompt.git \
  .claude/skills/realistic-video-prompting
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Realistic Video Prompting in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Realistic Video Prompting on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Realistic Video Prompting is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

真实视频提示词生成器

把用户的一个模糊想法("我想要一段韩国社区的日常感视频")转化为高密度、可执行、去AI感、像真实素材拍出来的结构化视频提示词。

核心哲学

高质量视频提示词的本质是回答一个问题:这段素材是谁、用什么设备、在什么年代、出于什么目的拍下来的?

一旦确定了"素材来源身份",所有细节都能推导出来。2005年朋友随手拍的DV会抖、会失焦、会突然关机;2024年博主的手机竖屏会有美颜感、会对着镜头说话;90年代的监控摄像头没有声音、俯视角、低帧率。

第二个关键认知:AI模型有强烈的默认审美(电影感运镜、完美构图、商业调色、干净画面)。想要真实感,必须用负面约束主动关闭这些默认行为,并在时间轴里植入"非完美事件"。

工作流程

第一步:确定四要素(缺一不可)

如果用户没有说清楚,先问这四个问题(可以合并成一次提问):

  1. 拍摄者身份:谁在拍?朋友、家人、路人、Vlogger、监控、纪录片摄制组、还是无人拍摄的固定机位?
  2. 设备与年代:DV、VHS、8mm胶片、iPhone、GoPro、监控、专业电影机?哪个年代?
  3. 主体与场景:拍谁?在哪?什么时间段?
  4. 情绪目标:温暖怀旧、孤独疏离、紧张不安、平静治愈?

如果用户已经给出足够信息,直接推断并在输出前用一句话确认你的理解。

第二步:按七段式模板逐段生成

严格按以下顺序输出,每段独立成块,段间不混淆信息:

主要角色 → 地点 → 视觉风格 → 摄像风格 → 时间轴 → 音频 → 目标

各段写作规则:

1. 主要角色

  • 只写可视化的具体特征:年龄段、妆容、每件衣物的颜色/材质/版型、鞋、配饰、发型细节(含碎发/刘海等小特征)
  • 禁用抽象形容词(时尚的、好看的、有气质的)
  • 加一句皮肤/质感描述(如"逼真的皮肤纹理")
  • 加一句性格气质(用于驱动肢体语言)
  • 必须以一致性锁定句结尾:"在整个视频中保持一致的身份、服装、发型和外貌"

2. 地点

  • 时间段 + 地域 + 场景类型开头
  • 罗列8到12个具体环境元素(盆栽、晾衣绳、电线杆、树影等),元素要能相互佐证同一个世界
  • 加光影动态描述(树影移动、云影掠过)
  • 以负面约束结尾:排除模型爱加的商业元素(商店、广告、人群、咖啡馆)

3. 视觉风格

  • 用4到6个短语定义真实感层级:如"超现实主义纪录片真实感、真实的即兴行为、自然的肢体语言、无剧本的日常生活片段感、强烈的环境真实性"

4. 摄像风格

  • 这是去AI感的核心段落。查阅 references/camera-aesthetics.md,选取对应设备的"缺陷参数包"完整写入
  • 原则:把设备美学翻译成6到10个具体物理现象(对焦搜索、曝光波动、滚动快门、压缩伪影等)
  • 必须以三连负面约束结尾:没有稳定、没有电影化摄像机移动、没有现代色彩分级(按设备调整)

5. 时间轴

  • 总长10到15秒,切5到7个镜头,每镜头2到3秒,用 00:00–00:02 格式
  • 每个镜头包含三层:主体动作 + 环境呼应(风、光、动物、路人)+ 一个非完美事件
  • 非完美事件从 references/anti-ai-checklist.md 的清单中选取,且7个镜头里不重复使用同一种
  • 动作之间要有生活逻辑链(坐着 → 走进小巷 → 遇到猫 → 晾衣服),别写成互不相关的镜头拼贴
  • 结尾必须有收束方式:突然切黑(像关机)、走出画面、镜头垂下拍到地面等,与拍摄者身份匹配

6. 音频

  • 先声明模式(仅自然环境音 / 现场收音 + 环境)
  • 罗列6到9个具体声音元素,与场景元素一一对应(有猫就有猫叫,有晾衣绳就有织物声)
  • 三连否定结尾:没有音乐、没有音效设计、没有旁白(按需调整)

7. 目标

  • 一句话灵魂锚点,格式:"捕捉XX,仿佛一段XX的XX,用4到6个形容词收尾"
  • 这句话是模型在细节冲突时的取舍依据,必须与前面所有段落自洽

第三步:自检(输出前必做)

对照 references/anti-ai-checklist.md 逐条检查:

  • 角色段有一致性锁定句吗?
  • 地点段排除了商业元素吗?
  • 摄像段有至少6个物理缺陷描述 + 3个"没有XX"吗?
  • 时间轴每个镜头都有一个非完美事件吗?有重复吗?
  • 结尾有收束方式吗?
  • 音频段的声音元素与画面元素对应吗?
  • 全文有没有漏进抽象形容词(美丽的、电影感的、高级的)?
  • 目标句与拍摄者身份/设备/年代自洽吗?

第四步:交付

  • 输出完整提示词(中文或英文按用户需求,投喂给模型时英文通常效果更稳,可主动提供英文版)
  • 附一段简短说明:这条提示词的"素材来源身份"设定是什么,用户想调整时改哪几个旋钮(设备包、时间段、情绪词、收束方式)

参考文件

  • references/camera-aesthetics.md:8种设备/媒介的缺陷参数包,写摄像风格段时必读
  • references/atmosphere-dictionary.md:氛围词到具体视觉现象的翻译表,写地点和时间轴时查阅
  • references/anti-ai-checklist.md:AI感来源清单、负面约束句库、非完美事件库,自检时必读

变体处理

  • 用户要非写实风格(动画、赛博朋克、广告片):七段式结构不变,但摄像段改为对应媒介的美学参数(动画帧率、广告片的运镜语言),负面约束改为排除该风格不要的元素
  • 用户要多角色:每个角色独立一段描述,各自带一致性锁定句,并在时间轴中明确角色间互动
  • 用户只要优化现有提示词:先按七段式诊断缺哪些段、哪些段密度不足,给出诊断表再重写
  • 超过15秒的长视频:建议拆成多条提示词分段生成,每条保持角色段完全一致以维持连续性

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Realistic Video Prompting AI skill do?

为AI视频生成模型(Seedance、Sora、Kling、Runway、Veo等)撰写真实感、去AI感、纪录片质感的结构化视频提示词。当用户想生成AI视频提示词、优化现有视频提示词、追求"真实感""家庭录像感""手机随拍感""DV感""监控录像感"的视频效果,或提到Seedance、文生视频、视频prompt时,务必使用此skill。即使用户只是描述了一个想拍的场景或氛围,也应使用此skill将其转化为完整提示词。

Why use Realistic Video Prompting on TypingMind?

Because you install it once and use it with any model. Realistic Video Prompting is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Realistic Video Prompting in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/zhouwei713/seedance-prompt/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Realistic Video Prompting?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Realistic Video Prompting?

As many as you like. As long as a model supports skills, you can use Realistic Video Prompting with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Realistic Video Prompting AI skill free?

Yes. It is published on GitHub by zhouwei713 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇