One Prompt Video logo

One Prompt Video

Community
huangbai-AI
one-prompt-video

一句话生成一条完整的口播视频:写稿 → 生成口播数字人(即梦 CLI / Seedance 2.5,本人照片 + 本人声音)→ 做 B-roll(截图、本人素材、代码动画)→ HyperFrames 合成 → 分段渲染 → 抽帧检查 → 按平台出版本。用户说“帮我做一条口播视频”“一句话出片”“按我的风格做期教程视频”“把这期做成成片”时使用。

Overview

Publisherhuangbai-AI
Repositoryone-prompt-video-skills
Skill nameone-prompt-video
Stars
96
Forks
10
Bundled files
4
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.

  • 4 bundled files

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

  • Open source

    Published by huangbai-AI on GitHub. Read the source before you install it.

Installation

Install the One Prompt Video 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/huangbai-AI/one-prompt-video-skills/tree/main/one-prompt-video
  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/huangbai-AI/one-prompt-video-skills.git /tmp/one-prompt-video-skills
mkdir -p .claude/skills
cp -r /tmp/one-prompt-video-skills/one-prompt-video .claude/skills/one-prompt-video
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable One Prompt Video 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 One Prompt Video 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 One Prompt Video 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.

一句话成片

把一条 1–3 分钟的口播教程视频从零做到可发布。人只需要给主题和素材、确认花钱的步骤、最后审片。

〇、开工前要拿到的东西

东西用途没有时
主题 + 目标观众 + 必须出现的事实写稿问一次
往期视频的标题 / 文案 / 链接学他的写稿风格用写稿 skill 里已有的风格
一张本人正脸照口播数字人的参考图最好从他真实视频里截一帧,不要用 AI 生成的人像
一段 8 秒左右干净的本人声音口播数字人的参考音频从往期视频里切一段没有配乐的
要发哪些平台决定出几个版本默认只出一个

花钱的步骤一律先问:每次提交生成前,列出模型、时长、预计积分,等对方明确说“提交”再提交;提交后报 submit_id 和余额。重跑也要问。

一、写稿

  1. 用写稿 skill(没有就先根据他往期视频总结一个)写三列分镜表:阶段|口播|画面。
  2. 每段口播 ≤ 15 秒;画面列标清【A】口播 /【B】B-roll /【B+A小窗】,B-roll 写明是截图、本人素材还是说明动画。
  3. 不编造数字;画面里不出现平台 logo、水印、网址。
  4. 分镜表给对方确认后再往下做。

二、口播画面(A-roll)

  1. 即梦 CLI 全能参考(multimodal2video),Seedance 2.5,16:9,参考图 + 参考音频,按分镜段落逐段生成。提示词见 references/prompts.md。
  2. 生成后逐段转文字(whisper),和台词逐字比对;念错的段落只重跑那一段(先报价再提交)。念错的词如果在全屏 B-roll 下面,可以直接换用别的 take 里那一句的声音,不必重跑。
  3. 下载原片保存,不要转码压缩后再用。

三、B-roll

  • 截图 / 录屏:真实页面、终端、文件夹。平台标识、账号、隐私打码。
  • 本人素材:往期视频片段、封面、照片,按原比例用,不裁切人脸。
  • 说明动画:用代码动画 skill;每个动画先定好落在哪句台词上(用字级时间轴的 cue),再逐帧画。一个画面只讲一件事。

四、口播粗剪(最容易出问题的一步)

用 ffmpeg 一次滤镜图完成(每段 trim → setpts → scale → fps=30,再 concat),规则:

  1. 只在停顿里下刀:剪点吸附到 ±0.25 秒内最安静的 20ms;停顿段必须在下一个字开口前 0.05 秒收住,否则会带进半个字。
  2. 倍速按帧率选:25 帧素材 1.2×、24 帧(Seedance)1.25×,正好映射到 30 帧,一帧不重复。1.1× 之类会周期性重复帧(跳帧)。
  3. 气口:每句话讲完留 0.5–1 秒;关键句(包袱、情绪句)后约 1 秒;笑声按原速放完再切。
  4. 气口不够长时慢放静默段,用 minterpolate 补帧;补帧会吃掉末尾几帧 → 多喂 0.25 秒原素材,再按精确帧数截断(tpad=stop_mode=clone,trim=end_frame=N,音频同样 apad + atrim 到对应采样数)。素材到头没法多喂时,别慢放,改用下一条素材开口前的静默当气口。
  5. 口播画面不做任何处理:不缩放(不做 100%/120% 跳切)、不推拉、不调色、不叠颗粒。720p 放大会发虚;两条不同 take 硬接的地方,用 B-roll 转场或小窗放大盖住。
  6. 缩放用 lanczos,中间文件 CRF ≤ 12,每秒一个关键帧(-g 30)。
  7. 粗剪后跑 scripts/check_frames.py:总帧数 = 时长 × 30;步骤段不允许出现 ≥2 帧的定格(章节卡除外)。

五、时间轴

  • whisper 字级对齐到成片口播,得到每句 s / e 和每个字的时间 ct;B-roll、字幕、动画都按 ct 取 cue。
  • 笑声、极短的词附近对齐常常错位。关键位置(要哔掉的词、硬切点)把那一小段单独切出来转文字确认,别直接信对齐结果。

六、HyperFrames 合成

  • 一个 index.html:口播 <video> + 若干插入片段 + 两个 canvas(背景 / 前景),一条暂停的 GSAP 时间线驱动 renderFrame(t),画面只由 t 决定(可逐帧导出、可跳到任意时间)。
  • 镜头表:每个镜头 = {模式 A / B / BP, 起点(取自台词 cue), 绘制函数, 转场}。B→A 快切;A↔A 不做动画;需要硬切时起点放在口播剪接点前一帧。
  • 字幕逐句、底部居中;配乐原创,人声出现时 sidechain 压低,整体 -14 LUFS。
  • 平台版本用一个开关(如 window.HB_VARIANT)切换:换图、打码(把文字画到离屏画布再缩小放大成马赛克)、换头像,主版本输出不受影响。

七、分段渲染

整片固定切成 N 段(如 8 段 × 16 秒),每段是一份把时钟起点设成段首的 html(口播加 data-media-start,插入片段平移或去掉,去掉音频),hyperframes render -c 渲完核对帧数,再用 concat 拼接,最后单独混音、封装。改哪段只重渲哪段;平台版本只渲它改到的段。脚本见 scripts/render_segments.py。

八、抽帧检查(交付前必须做完)

  • 每 2 秒抽一帧拼图,逐张看:文字出框、穿帮、黑帧、遮挡人脸、B-roll 与台词对不上
  • check_frames.py:帧数、剪接点后是否定格、每句之后的停顿长度
  • 全片转文字,和台词逐句核对;平台版本再搜一遍不能出现的词
  • 响度 -14 LUFS 左右、峰值 ≤ -1 dBFS
  • 开头结尾:最后一个字说完再淡出,不要压在字上

九、交付

  • 主视频(口播 + 转场音效)和 BGM(已避让人声 / 未避让)分开给,时长一致、从 0 对齐,叠加即原混音。
  • 平台版本:小红书把工具 / 模型名在画面里打码、口播里用 1 kHz 哔声替换;视频号等可换真人头像。改动涉及的每一处(包括缩略图、截图里的小头像)都要查到。
  • 附一份修改清单:每轮意见、改法、状态。

参考

  • references/prompts.md:写稿、口播生成、截图、代码动画、合成五段可直接复制的提示词
  • scripts/check_frames.py:帧数 / 定格 / 停顿检查 + 抽帧拼图
  • scripts/bleep.py:把指定时间段的人声换成哔声
  • scripts/render_segments.py:HyperFrames 分段渲染与拼接

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 One Prompt Video AI skill do?

一句话生成一条完整的口播视频:写稿 → 生成口播数字人(即梦 CLI / Seedance 2.5,本人照片 + 本人声音)→ 做 B-roll(截图、本人素材、代码动画)→ HyperFrames 合成 → 分段渲染 → 抽帧检查 → 按平台出版本。用户说“帮我做一条口播视频”“一句话出片”“按我的风格做期教程视频”“把这期做成成片”时使用。

Why use One Prompt Video on TypingMind?

Because you install it once and use it with any model. One Prompt Video 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 One Prompt Video in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/huangbai-AI/one-prompt-video-skills/tree/main/one-prompt-video. 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 One Prompt Video?

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 One Prompt Video?

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

Is the One Prompt Video AI skill free?

It is published on GitHub by huangbai-AI. Check the repository for licensing terms. 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 👇