npx skills add minimax-ai/minimax-h3 --skill brand-promo-video-generator
sjh00/minimax-h3-storyboard-prompt-skills
brand-promo-video-generator
品牌/产品宣传短片的 H3 提示词写法:品牌事实、叙事脊柱、精确节拍,输出可分段投喂 MiniMax H3 的 prompt。 适用发布/官网/社媒宣传提示词。不适用无授权仿标识、虚构卖点长剧本。
Installation
npx skills add sjh00/minimax-h3-storyboard-prompt-skills --skill brand-promo-video-generator
Similar popular skills
Related neighbors and high-traction skills in the same topics — useful to compare before installing.
Generate Excalidraw diagrams from natural language descriptions. Use when asked to "create a di…
29.1K installsGenerate a complete MCP server project in TypeScript using the MCP TypeScript SDK v2 (@modelcon…
12K installsComprehensive project architecture blueprint generator that analyzes codebases to create detail…
12K installsGenerate a complete MCP server project in Python with tools, resources, and proper configuration
10.3K installsIntelligent README.md generation prompt that analyzes project documentation structure and creat…
9.6K installsComprehensive technology-agnostic prompt for analyzing and documenting project folder structure…
9.4K installsAlso in this package
Other skills from sjh00/minimax-h3-storyboard-prompt-skills.
npx skills add sjh00/minimax-h3-storyboard-prompt-skills
More details
Agent compatibility
Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.
Also listed on
Alternate registries and mirrors of this skill.
npx skills add https://modelscope.cn/collections/MiniMax/MiniMax-H3-Skills
Repository health
main
Skill metadata
Parsed from SKILL.md frontmatter.
Package contents
Files included with this skill beyond the listing page.
-
skill md
SKILL.md2,295 B -
docs
SUMMARY.md275 B
History
- First seen on skills.sh
- First recorded snapshot · 43 installs
SKILL.md
品牌宣传 · 提示词写法
交付:节拍表(可中文)+ 分段 H3 提示词。 不做合成工程。
写前信息
- 可写进画面的素材事实:LOGO/产品/界面/主张/CTA(无授权则不仿造,标概念)
- 时长(常 15–30s)、画幅、受众、文案语言(跟品牌与平台,不机械跟聊天语)
叙事脊柱(写进每段风格句)
| 类型 | 脊柱 |
|---|---|
| SaaS/AI | 意图→规划→能力→执行→输出→证明→LOGO |
| 实体 | 英雄→交互→特写→场景→结果→LOGO |
| 服务 | 背景→流程→证据→成果→承诺→LOGO |
15s≈5–8 拍;30s≈8–12 拍。一拍一主动作。
节拍草稿
| 时间 | 视觉主导 | 动作 | 证明点 | 文案 | 意图(铺垫/冲击/制动) |
|---|
转场优先:产品动作、界面流、边缘匹配;忌假 HUD、未验证指标墙。
H3 终稿(每段 ≤15s)
[规格] {段时长}s · {画幅} · brand promo · H3
[风格] {品牌气质关键词} · 清晰信息层级 · 商业成片感
[本段节拍]
[Shot n] … 真实产品/界面描述(保真,不重绘 Logo)
文案:"…" 出现时段与停留
CTA / 主张仅用已确认表述
[衔接] 上段结束态 → 本段起态(多段时)
同色调、同字体气质、同光位基线
[声音] 原生:品牌安全器乐 + 轻 UI;写乐器速度动态
[负面] 假 Logo, 虚构功能, 杂乱字, 廉价模板, 水印
15s:多段 prompt + 衔接句;硬切落在动作完成或文案换句处。
压官方字段时接 h3-prompt-writing。
自检
- 主张与素材一致或标概念
- 身份资产未仿造
- 每段有视觉主导与证明点
- 文案语言符合投放