Menu
竖版大字报封面制作。第一眼只有一个大钩子，人物置顶，大字压在人物后面，底部一个结果承诺。 适合9:16短视频封面、真实案例、避坑内容、老板视角、方法论内容。 当用户说"做封面"、"短视频封面"、"大字报封面"、"竖版封面"、"封面图"时使用。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
横版口播内容增强剪辑。人物始终是主视觉，左右安全区放标题、步骤、对比、结果卡，组件克制不抢话。 适合横版口播课、工具教程、商业分析、方法论讲解、案例复盘。 当用户说"横版口播增强"、"横版讲解视频"、"横版内容增强剪辑"、"人物为主横版"时使用。
横版诊断盘开场口播增强剪辑。先用全屏主题判断盘钉住问题，再切人物建立信任。 适合账号置顶视频、课程第一课、服务定位说明、业务场景诊断、方法论总览、"第一次来先看这条"。 当用户说"诊断盘开场"、"置顶视频"、"账号介绍视频"、"横版开场增强"、"主题先行"时使用。
Create 16:9 horizontal kinetic talk-video edits with Remotion from a speaker video and script. Use when the user asks to cut or template a horizontal口播 video with 人物开场, full-screen animated business/methodology boards, phrase-triggered components, subtitles synced to voice, picture-in-picture circular speaker anchors, camera-like push/pan transitions, or the 横版运镜增强 style.
Use this skill when editing a 16:9 horizontal Chinese talking-head video with Codex. It guides source inspection, script-aligned captions, scene segmentation, full-screen speaker moments, large material boards, picture-in-picture speaker anchors, phrase-triggered animation, Remotion rendering, and pre-delivery quality checks.
Use this skill when editing a 9:16 vertical Chinese talking-head video with Codex. It guides short-video opening hooks, face-safe layouts, large subtitles, phrase-triggered cards, material cut-ins, speaker PIP, vertical Remotion rendering, and mobile-first quality checks.
| name | vertical-big-word-cover |
| description | 竖版大字报封面制作。第一眼只有一个大钩子，人物置顶，大字压在人物后面，底部一个结果承诺。 适合9:16短视频封面、真实案例、避坑内容、老板视角、方法论内容。 当用户说"做封面"、"短视频封面"、"大字报封面"、"竖版封面"、"封面图"时使用。 |
Create a 9:16 short-video cover that works as an attention interceptor, not an information page. The viewer should see one hook, one face, one result promise — nothing more.
A cover is not an explanation page. It is an attention interceptor.
The shared trait of effective covers is NOT "lots of text" — it is:
Person must cover the big text, never the reverse.
Extract "conflict + action + result", not "asset name":
Conflict: 别做X / 别问X / X不等于Y
Action: 先做Y / 先拆Y / 先跑通Y
Result: 变成Z / N天见效 / 3步搞定Z
Bad example: "一堆资料" — just an asset, no relation to the outcome. Good example: "别问需求 / 先拆资料 / 一堆文档变AI工作流"
Three layers maximum:
Never show simultaneously: multiple labels, multiple explanation sentences, multiple process nodes, multiple peer-level titles.
别做系统
先跑场景
7天见效
For: enterprise AI, methodology, pitfall-avoidance content.
一堆资料
变AI工作流
真实案例
For: customer cases, project reviews, data governance content.
会用AI
不等于提效
先选场景
For: cognitive correction, boss perspective, team efficiency content.
Fail if:
Read references/cover-sop.md for the full SOP with detailed layer specs, copy extraction methods, and finished cover examples.