Ace Step 1.5 Prompt Writer logo

Ace Step 1.5 Prompt Writer

Community
swan7-py
ace-step-1.5-prompt-writer

通过互动式问答了解用户想要的音乐(用途/情绪/风格/乐器/人声/歌词/元数据),最终生成 ACE-Step 1.5(本地开源音乐模型)的生成提示词:Caption + Lyrics + Metadata 三段式。纯提示词,不调用任何生成接口。供 8 个本地版视频 skill 产出 BGM,或用户直接要 ACE-Step 1.5 提示词时使用。

Overview

Publisherswan7-py
RepositoryMiniMax-H3-Skills-Local
Skill nameace-step-1.5-prompt-writer
Stars
66
Forks
5
Bundled files
1
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.

  • 1 bundled files

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

  • Open source

    Published by swan7-py on GitHub. Read the source before you install it.

Installation

Install the Ace Step 1.5 Prompt Writer 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/swan7-py/MiniMax-H3-Skills-Local/tree/main/old/ace-step-1.5-prompt-writer
  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/swan7-py/MiniMax-H3-Skills-Local.git /tmp/MiniMax-H3-Skills-Local
mkdir -p .claude/skills
cp -r /tmp/MiniMax-H3-Skills-Local/old/ace-step-1.5-prompt-writer .claude/skills/ace-step-1.5-prompt-writer
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ace Step 1.5 Prompt Writer 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 Ace Step 1.5 Prompt Writer 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 Ace Step 1.5 Prompt Writer 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.

ACE-Step 1.5 提示词书写 skill(本地版 · 互动式 · 纯提示词)

把用户的音乐想法,通过多轮互动问答逐步澄清,最终改写为 ACE-Step 1.5 的生成提示词(Caption + Lyrics + Metadata 三段)。本 skill 只产出文本提示词,不调用任何生成接口,由用户在本地 ACE-Step 1.5(Gradio UI / REST API / ComfyUI 节点)中自行执行。

何时调用

  • 用户说「写一段 ACE-Step 1.5 的 BGM / 写首歌的提示词 / 帮我想个配乐 / 配个背景音乐」。
  • 8 个视频本地版 skill 在「音频模式门 = 分离生成」时,需要单独产出 BGM / 整曲提示词。
  • 任何需要 ACE-Step 1.5 caption / lyrics / 元数据文本的场景。

核心互动流程(逐步确认,别一次问完所有问题)

遵循「先大类后细节、能选就给选项」的原则。用 AskUserQuestion 一次性抛 2–3 个相关问题(每个选项 3–4 个 + 用户可填「其他」),解释文字要短。每轮确认后再进下一轮。最少两轮即可出结果,信息越足越精准。

用户若一开始就给了一段完整描述,可跳过提问直接进「收敛|生成输出」;若描述含糊,仍按流程补问缺口。

第一轮|用途 + 情绪 + 流派(必问)

用 AskUserQuestion 同时问这三件:

  1. 用途(决定人声与否与结构):
    • BGM 纯音乐(循环友好、无人声、用 [Instrumental])
    • 带人声整曲(有主歌副歌、有演唱)
    • 你来推荐(按后续情绪/场景判断,默认先给 BGM 方案)
  2. 情绪/氛围(给常见预设,用户可多选或填其他):
    • 科技感/未来感、温暖/治愈、紧张/悬疑、欢快/活力、梦幻/空灵、沉重/悲伤、史诗/宏大、复古/怀旧
  3. 流派(可选「让模型自由发挥」):
    • pop、rock、electronic/EDM、lo-fi、synthwave、jazz、classical、hip-hop/R&B、folk、cinematic、民乐

第二轮|乐器 / 音色 + 参考(按需)

  • 主要乐器:piano / acoustic guitar / synth pads / strings / 808 drums / electric bass / brass / 古筝·二胡等民乐 / choir
  • 音色质感:warm / bright / crisp / airy / punchy / lush / raw / polished
  • 时代/参考(可选):80s synth-pop / 90s grunge / 2010s EDM / vintage soul / modern trap / 某位艺术家(用 reminiscent of ... 或 in the style of ...)
  • ⚠️ 若用户给了参考艺术家/歌曲,优先用 in the style of / reminiscent of 写进 Caption,比堆形容词更高效、更稳定。

第三轮|人声与歌词

  • 无人声(BGM):Lyrics 直接写 [Instrumental],Metadata instrumental: true。
  • 有人声:
    • 语种:中文 / 英文 / 日文 / 其他(影响 vocal_language)
    • 人声特点:female/male vocal、breathy / powerful / falsetto / raspy / choir
    • 歌词来源: ① 用户提供文本 → 你帮忙切分并加 [Intro]/[Verse]/[Chorus] 等结构标记; ② 你代写 → 按下方 Lyrics 指南创作(带结构标记、每行 6–10 音节、坚持一个核心隐喻)。

第四轮|元数据(可选,最简可全交自动)

确认目标时长、BPM、调性、拍号。用户不给则标注「自动推断」,让 thinking 模式(use_cot_metas=true)自行推断——日常不必手写。

收敛|生成输出

汇总为下方「输出契约」三段式,直接交用户粘贴进 ACE-Step 1.5。若用户只给了模糊方向,先给一版再邀迭代(结尾提示「不确定可改:…」)。

Caption 写作指南(最重要输入,≤512 字符)

Caption 决定生成音乐的「整体画像」,支持三种形式:简单风格词、逗号分隔 tags、复杂自然语言——形式不影响效果,关键是维度齐全。 组合越多维度越精准(单一维度会让模型自由发挥):

维度示例词
风格/流派pop, rock, jazz, electronic, hip-hop, R&B, folk, classical, lo-fi, synthwave
情绪/氛围melancholic, uplifting, energetic, dreamy, dark, nostalgic, euphoric, intimate
乐器acoustic guitar, piano, synth pads, 808 drums, strings, brass, electric bass
音色质感warm, bright, crisp, muddy, airy, punchy, lush, raw, polished
时代参考80s synth-pop, 90s grunge, 2010s EDM, vintage soul, modern trap
制作风格lo-fi, high-fidelity, live recording, studio-polished, bedroom pop
人声特点female vocal, male vocal, breathy, powerful, falsetto, raspy, choir
速度/节奏slow tempo, mid-tempo, fast-paced, groovy, driving, laid-back
结构提示building intro, catchy chorus, dramatic bridge, fade-out ending

原则

  1. 具体优于模糊:「sad piano ballad with female breathy vocal」远胜「a sad song」。
  2. 组合多个维度(风格+情绪+乐器+音色)锚定方向。
  3. 善用参考:「in the style of 80s synthwave」「reminiscent of Bon Iver」快速传达复杂审美。
  4. 质感词(warm/crisp/airy/punchy)影响混音与音色倾向,很有用。
  5. 描述粒度决定自由度:写少→惊喜多;写细→可控强。
  6. 避免冲突词:如同时「古典弦乐」+「硬核金属」易劣化。解决:① 重复强化想要的元素;② 把冲突改写为时间演变(「开头柔和弦乐,中段噪杂金属摇滚,结尾转 hip-hop」)。

⚠️ 不要把 BPM / 调性 / 拍号写进 Caption——这些交给下方 Metadata 参数控制。Caption 专注风格/情绪/乐器/音色。

Lyrics 写作指南(时间脚本,≤4096 字符;纯音乐写 [Instrumental])

Lyrics 控制音乐随时间展开。结构标记可用 - 组合(如 [Chorus - anthemic]),不要堆叠太多标记(避免 [Chorus - anthemic - stacked harmonies - high energy])。

类别标记
基础结构[Intro] [Verse]/[Verse 1] [Pre-Chorus] [Chorus] [Bridge] [Outro]
动态段落[Build] [Drop] [Breakdown]
器乐段落[Instrumental] [Guitar Solo] [Piano Interlude]
特殊[Fade Out] [Silence]

技巧

  • 每行 6–10 个音节效果最好(模型把音节对齐到节拍)。
  • 大写 = 更强演唱力度(WE ARE THE CHAMPIONS!)。
  • 括号 = 背景人声(We rise together (together))。
  • 重复元音延音(Feeeling so aliiive,谨慎用)。
  • 段落间空行分隔。
  • 人声控制可叠加:[raspy vocal] [whispered] [falsetto] [powerful belting] [spoken word] [harmonies] [ad-lib]
  • 能量/情绪可叠加:[high energy] [building energy] [explosive] [melancholic] [euphoric] [dreamy]

避免「AI 味」歌词:形容词堆砌、押韵混乱、段落边界模糊、没有呼吸感(每行太长)、隐喻混用(一首歌坚持一个核心隐喻)。

⚠️ 保持 Caption 与 Lyrics 一致:Caption 写「violin solo, classical」则 Lyrics 用 [Violin Solo - expressive],不要冲突成 [Guitar Solo - electric]。

Metadata 指南(按需填写;不填则开 thinking 模式由 LM 自动推断)

参数范围说明
bpm30–300速度;慢歌 60–80、中速 90–120、快歌 130–180
keyscale调性如 C Major Am F# Minor
timesignature拍号4/4(最常见) 3/4(华尔兹) 6/8(摇摆)
vocal_language语种码zh en ja es fr…(LM 通常能据歌词识别)
duration秒10–600;实际略有偏差
instrumentaltrue/falsetrue 则无论歌词都生成纯音乐

输出契约(固定三段,用户可直接粘贴进 ACE-Step 1.5)

### Caption(≤512字符,专注风格/情绪/乐器/音色,不写BPM/调性)
<组合维度后的描述文本>

### Lyrics(≤4096字符;纯音乐写 [Instrumental])
[Intro - piano]

[Verse 1]
...

[Chorus - powerful]
...

[Outro - fade out]

### Metadata(不填的项标注「自动推断」)
- bpm: 120
- keyscale: C Major
- timesignature: 4/4
- vocal_language: zh
- duration: 30
- instrumental: false

示例

示例 1|科技感 BGM(纯音乐,循环友好)

### Caption
electronic, mid-tempo, tech-house, warm analog synth pads, punchy kick, airy pluck arpeggio, studio-polished, building intro, subtle tension

### Lyrics
[Instrumental]

### Metadata
- bpm: 100
- keyscale: A Minor
- timesignature: 4/4
- vocal_language: (自动推断)
- duration: 30
- instrumental: true

示例 2|中文人声宣传曲

### Caption
pop, uplifting, energetic, bright, piano and strings, powerful female vocal, stadium chorus, studio-polished

### Lyrics
[Intro - piano]

[Verse 1]
迎着晨光出发
心跳和鼓点同频
每一步都算数
这是我们自己的路

[Chorus - powerful]
向前跑 不回头
光在掌心燃烧
这一刻 属于我们
WE ARE THE LIGHT!

[Outro - fade out]

### Metadata
- bpm: 120
- keyscale: C Major
- timesignature: 4/4
- vocal_language: zh
- duration: 45
- instrumental: false

示例 3|参考风格(用参考词快速传达审美)

### Caption
lo-fi, dreamy, nostalgic, warm vinyl crackle, soft electric piano, breathy male vocal, reminiscent of 90s city-pop, laid-back

### Lyrics
[Verse 1]
夜色漫过窗台
老唱片轻轻转
...

[Chorus]
就让它慢慢走
像潮水不退不进

[Outro - fade out]

### Metadata
- bpm: 85
- keyscale: F Major
- timesignature: 4/4
- vocal_language: zh
- duration: 40
- instrumental: false

互动对话示例(供 agent 参考语气)

Agent: 想做哪种音乐?(用 AskUserQuestion 抛:用途 / 情绪 / 流派)
User:  要个 BGM,科技感,electronic
Agent: 乐器倾向?(synth pads / piano / 弦乐…)有参考艺术家吗?
User:  synth pads + 808 鼓,类似 80s synthwave
Agent: 时长要多长?BPM 有要求吗?
User:  30 秒循环,其他你定
Agent: (输出示例 1 风格的三段式,结尾提示「想改情绪/加人声可说」)

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 Ace Step 1.5 Prompt Writer AI skill do?

通过互动式问答了解用户想要的音乐(用途/情绪/风格/乐器/人声/歌词/元数据),最终生成 ACE-Step 1.5(本地开源音乐模型)的生成提示词:Caption + Lyrics + Metadata 三段式。纯提示词,不调用任何生成接口。供 8 个本地版视频 skill 产出 BGM,或用户直接要 ACE-Step 1.5 提示词时使用。

Why use Ace Step 1.5 Prompt Writer on TypingMind?

Because you install it once and use it with any model. Ace Step 1.5 Prompt Writer 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 Ace Step 1.5 Prompt Writer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/swan7-py/MiniMax-H3-Skills-Local/tree/main/old/ace-step-1.5-prompt-writer. 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 Ace Step 1.5 Prompt Writer?

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 Ace Step 1.5 Prompt Writer?

As many as you like. As long as a model supports skills, you can use Ace Step 1.5 Prompt Writer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Ace Step 1.5 Prompt Writer AI skill free?

It is published on GitHub by swan7-py. 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 👇