Ancient Rap Battle

Community

10291124jcjc-dotcom

ancient-rap-battle

古人说唱 Battle：把两个历史上的死对头（诸葛亮 vs 司马懿、岳飞 vs 秦桧……）做成一条 AI 说唱对决短视频，从选题写词（每句配史料注释）、Suno 做歌、Midjourney 定妆、即梦 Seedance 2.5 网页版分段生成（提示词按半句精确到 0.01 秒写歌词，口型一次对好），到自动逐镜头对齐、拼接、字幕包装、封面和发布文案一条龙。用户说"做一期 A vs B"、"古人说唱 battle"、"历史人物说唱"、"rap battle 视频"，或者要改已有一期的歌词、镜头、字幕、封面时使用。Make an AI rap-battle short between two historical rivals, end to end, with Seedance 2.5 (or MiniMax H3).

Overview

Publisher	10291124jcjc-dotcom
Repository	`ancient-rap-battle`
Skill name	`ancient-rap-battle`
Stars	133
Forks	28
Bundled files	28
License	MIT
Links	Skill source Repository

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.
28 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by 10291124jcjc-dotcom on GitHub. Read the source before you install it.

Installation

Install the Ancient Rap Battle AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

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.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/10291124jcjc-dotcom/ancient-rap-battle/tree/main
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.

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/10291124jcjc-dotcom/ancient-rap-battle.git \
  .claude/skills/ancient-rap-battle

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ancient Rap Battle 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 Ancient Rap Battle 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 Ancient Rap Battle 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.

古人说唱 Battle

两个古人在擂台上当面 diss。成片约 90 秒、16:9、2K，前奏 2.5 秒就开唱，两边各 16 句，每句底下一行史料出处。

0. 先说清楚的三件事

用户要有：Suno 账号、Midjourney 账号、即梦会员（Seedance 2.5 网页版，按积分）。可选：MiniMax API Key 走 H3 路线
红线：不用近现代政治人物、不用在世真人；脸不能像任何真实的人或演员。请求涉及这些时礼貌拒绝，建议换成古人
不碰别人的作品：不帮去别人视频的水印、不搬运；拆解学习可以

全程用中文和用户说话。API Key 只通过环境变量传入，绝不写进文件或记忆。

工作区

每期一个文件夹（下面叫 EP），先 python S/setup_check.py（S/ = 本 Skill 的 scripts/，Windows 用 python）。


EP/
  episode.json      ← 本期设定（从 examples/zhugeliang-vs-simayi/episode.json 复制改）
  01_歌词.md         ← 给用户看的歌词 + 注释 + 本期改动
  song.mp3  refs/*.png  音乐_最终.wav  字幕.srt
  seedance/NN_.../  ← 每段的素材包，用户把生成的视频下载回各自文件夹
  work/  preview/  final/  cover/

流程

第 1 步：选题与歌词（Claude 写，用户确认）

按 references/lyrics-guide.md：赢家后唱；两边各 16 句、严格四句一换韵；一句一个典故，句中用空格分前后半句；后唱的接住前面 3–4 句原话；每句一个照字面演的动作；每句一条史料注释（正史引原文，演义标明）。写之前读往期歌词查重，同一人物不复用旧梗旧句式。

先给用户纯歌词，再给注释，再给「本期改动」。用户挑出像旧梗的句子就换。定稿写进 episode.json（lyrics、notes、shots、look_cn、listen_cn、venue_cn）。

第 2 步：音乐（用户在 Suno 做）

按 references/suno.md：Advanced、Variety 拉到 0；A 一段生成、挑好后 Extend 接 B 一段（Extend from 设在 A 最后一句唱完处），Get Whole Song 下载 MP3 存成 EP/song.mp3。

第 3 步：歌曲分析与剪短


python S/analyze_song.py EP

看 line check（32 句都唱到）。然后用 whisper 词时间查间奏：Suno 常在口号后自己连唱一段 B 的第一句。剪间奏用 retrim.py，手法见 pitfalls.md「间奏拼接」——用户不接受任何停顿或卡顿：


python S/retrim.py EP --keep 7.36-53.64,63.63-109.8 [--mute ...] [--set b1=62.25]

把 音乐_最终.wav 发给用户听（接口、两人声音分不分得开；不开就 pitch 各 ±1 半音后重跑）。

第 4 步：定妆图（用户在 Midjourney 做）

按 references/midjourney.md 给 9 条提示词：两人各三视图 / 表情四宫格 / 半身（A 朝右、B 朝左），DJ、观众、场景（正面对称）。不做气氛组。逐张看图，复制改名进 EP/refs/。

第 5 步：素材包


python S/clean_vocal.py EP        # 更干净的人声，只给对口型用
python S/seedance_packs.py EP     # 生成 seedance/00_开场 … 10_片尾

每个唱歌的包：人声（从第一个字到下一句开口前，前补 0.3 秒静音）、半身 / 场景 / DJ / 表情图、提示词。提示词里每个镜头写清「X.XX–X.XX 秒唱「半句」」——这是口型一次对好的关键（用户确认，以后每期都这样）。检查每个包的音频首尾字完整。

第 6 步：用户在即梦生成（按 `references/seedance.md`）

先 480p / 720p，按编号上传图片和音频、粘贴提示词、时长选文件里写的秒数。下载回对应文件夹。满意后用「超清」放大同一条成 2K，存成 2k.mp4。

第 7 步：对齐（每段生成完就跑）


python S/seedance_align.py EP 03        # 逐镜头对齐，出 preview/03_..._配音乐.mp4
python S/seedance_align.py EP 04 --whole # 用户说「这段没事，按 11 秒对」时

把配音乐的小样发给用户看。用户的眼睛是标准：嘴不对 → 重做那段，或换成对方的反应镜头；画面和歌词对不上 → 先问是不是贴错了提示词。

第 8 步：拼接与包装


python S/seedance_assemble.py EP --2k            # preview/成片_拼接.mp4（2K 60 帧）
python S/package.py EP --test 0:12                # 先出一小段看字幕
python S/package.py EP                            # final/<标题>_成片.mp4

包装：开场标题叠在推镜头上、金色字幕（最后两字放大垫红笔刷）、底部史料注释（特粗纯白）、名字卡（A 左 B 右）、片尾提问、水印、调色 B（去胶质感）、开场 / 间奏 / 片尾的现场声。字幕颜色大小、注释、调色都只改 package.py 重新导出，不用重新生成。

第 9 步：封面与文案


python S/cover.py EP --frame <一帧两人对峙的截图>

竖版 3:4 + 横版 16:9。文案：IG（中英双语）、YouTube Shorts（标题中英名字）、抖音 / 小红书（金句开头）+ 话题 + 结尾提问 + 置顶评论投票下期。海外平台（IG / YouTube）优先。

发布前提醒（每期都说）

平台勾选「AI 生成内容」，片头的「AI生成仅供娱乐」不要去掉
Suno 免费版的歌不能商用
注释出处定稿前核一遍原文

H3 API 路线（可选）

用户有 MiniMax 余额、想批量自动提交时：make_clips.py → h3run.py --estimate（先报价，同意了才提交）→ qa.py / lipcheck.py → assemble.py → package.py。模板见 references/shot-template.md 下半部分。

参考

references/lyrics-guide.md：选题清单、查重、歌词结构、注释、episode.json 字段
references/suno.md：设置、两段 Extend、风格模板
references/midjourney.md：9 张定妆图
references/seedance.md：即梦网页版生成、口型、清晰度
references/shot-template.md：整片节奏；H3 的 35 条模板
references/pitfalls.md：踩过的坑
examples/zhugeliang-vs-simayi/episode.json：Seedance 路线完整示例；examples/zhouyu-vs-zhugeliang/：H3 路线示例

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 Ancient Rap Battle AI skill do?

Why use Ancient Rap Battle on TypingMind?

Because you install it once and use it with any model. Ancient Rap Battle 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 Ancient Rap Battle in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/10291124jcjc-dotcom/ancient-rap-battle/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 Ancient Rap Battle?

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 Ancient Rap Battle?

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

Is the Ancient Rap Battle AI skill free?

Yes. It is published on GitHub by 10291124jcjc-dotcom 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

Comfyui

NousResearch

OrganizationPopular

Generate images, video, and audio via diffusion workflows.

Heartmula

HeartMuLa: Suno-like song generation from lyrics + tags.

Hyperframes

Render MP4/WebM videos from HTML compositions.

Kanban Video Orchestrator

NousResearch

OrganizationPopular

Plan and run multi-agent video production pipelines.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

古人说唱 Battle

0. 先说清楚的三件事

工作区

流程

第 1 步：选题与歌词（Claude 写，用户确认）

第 2 步：音乐（用户在 Suno 做）

第 3 步：歌曲分析与剪短

第 4 步：定妆图（用户在 Midjourney 做）

第 5 步：素材包

第 6 步：用户在即梦生成（按 references/seedance.md）

第 7 步：对齐（每段生成完就跑）

第 8 步：拼接与包装

第 9 步：封面与文案

发布前提醒（每期都说）

H3 API 路线（可选）

参考

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now

第 6 步：用户在即梦生成（按 `references/seedance.md`）