Skip to content
cccskills
FreePublished on GitHub

selling-point-storyboard

卖点分镜表:把商品卖点拆成 15–60 秒带货短视频的逐镜分镜表(时间码、画面、台词、字幕、花字、素材来源、运镜),并直接输出可渲染的 storyboard.json。

Use when the user wants a storyboard, 分镜, 分镜脚本, shot list, 带货视频脚本, video script from selling points, or needs to plan a product video before generating or editing.

The English summary is being prepared. Showing the original description.

Run score Not tested

See how to install

Files included (1)

  • SKILL.md4.9 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

卖点分镜表 · Selling-Point Storyboard

从「卖点清单」到「能直接渲染的分镜表」。输出两份:人看的 Markdown 表格 + 机器跑的 storyboard.json(给 ffmpeg-auto-assemble)。

Part of ecommerce-video-skills

输入 Inputs

  • 商品信息与 真实 卖点(3 个最好,最多 5 个)
  • 时长目标:15s(种草/主图视频)· 30s(标准带货)· 60s(测评/讲解)
  • 现有素材:商品图、场景图、实拍视频(文件名列出来)
  • 平台 + 画幅;配音语言与音色偏好

结构公式 Structure

时长结构(镜数)
15s钩子(1) → 产品亮相(1) → 卖点×2–3(2–3) → 行动号召(1),共 5–6 镜
30s钩子 → 痛点放大 → 产品亮相 → 卖点×3(每个「说 + 证明」)→ 使用场景 → 行动号召,8–10 镜
60s钩子 → 我的经历 → 开箱 → 卖点×3–4 演示 → 对比 → 缺点/适合谁 → 行动号召,14–18 镜

规则:一镜一信息;每个卖点都要配「证明画面」(演示、细节、对比),光说不演不算卖点。

步骤 Steps

  1. 卖点排序:按「用户最在意 × 画面最好证明」排序,最强的放第 2–3 镜。
  2. 每镜写:时间、画面(用哪张素材/要拍什么)、台词(每镜 ≤ 2 句,中文 4–5 字/秒)、字幕(关键词用 **加粗** 标高亮)、花字(可选)、运镜。
  3. 时长核算:台词字数 ÷ 4.5 ≈ 秒数;总时长超了就删台词,不要加快语速超过 +20%。
  4. 缺素材的镜头标记 TODO,并给出拍摄/生成建议(可交给 image-to-video-shots 或姊妹仓库 ecommerce-image-skills 出图)。
  5. 输出 Markdown 表 + storyboard.json。

分镜表模板

#时间画面 / 素材台词(配音)字幕(高亮)花字运镜
10–3slifestyle.webp 户外手拿杯夏天带冰水出门,不到中午就变温了?冰水不到中午就变温了?冰水撑不到中午?punch
23–5swhite-bg.webp 白底亮相换这只保温杯试试。换这只保温杯试试—zoom_in
…

storyboard.json 模板

{
  "voice": "zh-CN-YunxiNeural", "rate": "+10%", "bgm": "auto", "tag": "品牌名 · 型号",
  "shots": [
    {"image": "assets/scene.jpg", "motion": "punch", "line": "钩子台词", "sub": "钩子**关键词**", "title": "花字"},
    {"image": "assets/main.jpg", "motion": "zoom_in", "line": "产品亮相台词"},
    {"image": "assets/detail.jpg", "motion": "pan_down", "fit": "blur", "line": "卖点一台词", "sub": "卖点一|**证明**"},
    {"video": "clips/demo.mp4", "start": 3.0, "line": "卖点二演示台词"},
    {"image": "assets/end-card.png", "min": 3.0, "line": "行动号召台词", "sub": "", "cta": "点击下方链接 ›"}
  ]
}

字段说明见 ffmpeg-auto-assemble。渲染:python skills/ffmpeg-auto-assemble/scripts/assemble.py storyboard.json -o out/video.mp4。

提示词模板 Prompt template

你是电商短视频编导。根据以下信息写一份 {15} 秒竖屏带货分镜表。
商品:{…};目标人群:{…};平台:{…}
真实卖点(不得编造数据、认证、价格):{…}
可用素材文件:{lifestyle.webp, white-bg.webp, …}
要求:一镜一信息;每个卖点配证明画面;台词每镜 ≤ 2 句,总字数 ≤ {时长×4.5};
关键词用 **加粗**;最后一镜是行动号召。
先输出 Markdown 分镜表,再输出符合下面 schema 的 storyboard.json:{粘贴上方模板}

检查清单 Checklist

  • 第 1 镜是钩子(见 hook-3s-script),最后 1 镜是行动号召
  • 总台词字数 ≈ 时长 × 4.5(中文)/ × 2.5 词(英文)
  • 每个卖点都有证明画面;没有编造的数据、认证、价格
  • 每镜素材都存在或标了 TODO
  • JSON 能通过 python -m json.tool storyboard.json

常见问题与修复 Failure fixes

问题修复
成片超时删台词而不是加速;合并信息相近的镜头
卖点像念说明书每个卖点改成「场景 + 结果」:「放包里一下午,本子还是干的」
素材不够同一张图用不同 motion / fit 做两镜;或生成场景图
字幕挡商品卖点镜头用 "fit": "blur",商品放在上半屏

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

AI 口播配音:用免费的 edge-tts(微软 Edge 在线语音)为带货短视频逐句生成配音,自动去首尾静音、统计每句时长、缓存复用,可选音色/语速/音调,并给出口播稿改写规则。Use when the user needs a voice-over, 配音, 口播, 旁白, TTS, text-to-speech, AI 主播声音, or wants to turn a script into audio for 抖音 / TikTok / 视频号 videos.

The English summary is being prepared. Showing the original description.

xianyu110/ecommerce-video-skills97Updated Oct 5, 2026

封面与标题 A/B:为带货短视频生成多套封面(自动挑选清晰帧或用干净素材图 + 不同标题与版式)和多组标题/文案,导出 3:4 / 9:16 封面文件和对比图,并给出 A/B 测试方法与记录表。Use when the user wants a video cover, 封面, 首图, thumbnail, 标题, 文案, title variants, A/B test covers for 抖音 / 视频号 / 小红书 / TikTok, or wants to improve click-through rate.

The English summary is being prepared. Showing the original description.

xianyu110/ecommerce-video-skills97Updated Oct 5, 2026

ffmpeg 自动拼接成片:一条命令把 storyboard.json(商品图/视频片段 + 口播稿)渲染成带配音、字幕花字、转场、BGM 自动闪避、响度标准化的 9:16 MP4,纯本地运行。Use when the user wants to assemble/edit/render a product video, 自动剪辑, 一键成片, 拼接视频, 图片转视频, Ken Burns, add BGM/subtitles/voice-over with ffmpeg, or batch-produce short videos without an editor.

The English summary is being prepared. Showing the original description.

xianyu110/ecommerce-video-skills97Updated Oct 5, 2026

3 秒钩子脚本:为电商短视频写前 3 秒开场(台词 + 画面 + 花字),按痛点/反差/结果先行/提问/数字/场景代入等钩子类型一次给出多版,并做合规自检。Use when the user asks for a hook, 开头, 黄金 3 秒, 前三秒, opening line, scroll-stopper, 完播率, or wants to rewrite a weak opening for 抖音 / TikTok Shop / 视频号 / Reels.

The English summary is being prepared. Showing the original description.

xianyu110/ecommerce-video-skills97Updated Oct 5, 2026

主图转视频:把一张或几张商品主图/白底图/场景图,写成可直接喂给图生视频模型的镜头提示词(运镜、主体动作、光线、时长、负面约束),并保证商品不变形。Use when the user has product images and wants a product video, 主图视频, 图生视频, image-to-video, animate a product photo, or 让主图动起来 — with any image-to-video model (Seedance / 可灵 Kling / Veo / Runway / Hailuo / Wan …) or a local Ken Burns fallback.

The English summary is being prepared. Showing the original description.

xianyu110/ecommerce-video-skills97Updated Oct 5, 2026

直播切片:从电商直播录屏中找出高光片段(按时间戳 / 按字幕关键词 / 按停顿自动提议),批量导出 9:16 竖屏切片,自动模糊填充横屏画面、烧录标题与字幕、响度标准化,并给切片写标题与开头钩子。Use when the user has a livestream recording and wants 直播切片, 切片, highlights, clips, 高光片段, repurpose a live stream into short videos for 抖音 / 视频号 / TikTok.

The English summary is being prepared. Showing the original description.

xianyu110/ecommerce-video-skills97Updated Oct 5, 2026

See all skills by xianyu110

Report a problem with this skill