World as 100 people (icons + stat race + punchline)
Template (assume duration 10 unless scaled): 0–2s shrink the world to 100 people; 2–9s rapid visual changes (wealth split, gender, access to water/internet, etc.—user picks themes); 9–10s one shocking or memorable stat line (user-themed). English on-screen text with exact windows. Default duration is 10 when unspecified.
Dependencies: WERYAI_API_KEY + Node.js 18+. scripts/video_gen.js + WERYAI_VIDEO_API.md. No other Cursor skills required.
Prerequisites
- -
WERYAI_API_KEY must be set; Node 18+. Prefer public https image URLs; if the assembled scripts/video_gen.js supports local file paths, review/verify the script and explicitly consent before local read-and-upload to WeryAI. - Non-empty
model in JSON.
Security, secrets, and API hosts
- -
WERYAI_API_KEY: Secret; never commit. - API hosts fixed in
video_gen.js as in sibling skills—only the key is read from env. - Local image handling disclosure: Prefer public
https image URLs. If the assembled scripts/video_gen.js supports local file paths, it may read a local image and upload it to WeryAI to obtain a public URL; require review / verification and explicit consent before using that path. - Higher assurance: Prefer a short-lived or isolated environment; review
scripts/video_gen.js before production use. Verify whether the runtime can read local image files and upload them to WeryAI, and obtain explicit consent before using that path.
Prompt expansion (mandatory)
- - Duration: Default 10; 5 or 15 when allowed and requested.
- Timeline scaling: Rescale hook / middle / punchline segments proportionally; keep hook → three middle beats → final punchline order.
Locked rhythm (example for duration 10):
- - 0.0–2.0s — Hook: e.g. SHRINK EARTH TO 100 PEOPLE.
- 2.0–4.0s, 4.0–6.0s, 6.0–9.0s — Three rotating stat / theme lines (user brief): e.g. WEALTH SHIFTS / GENDER / WHO GETS WHAT—keep labels short; pair each with icon morph, pie wipe, or counter tick.
- 9.0–10.0s — Punchline stat (one short English line the user intends—e.g. ONLY X HAVE Y).
Visuals: Isometric or flat UI icons, glow, snappy transitions (~1s); no tiny unreadable numbers unless the caption carries the readable stat.
Typography: Bold sans, outline, lower third; English only unless user requests otherwise.
Audio: Light digital ticks / whoosh if audio on.
Negatives (Kling): garbled numbers, illegible text, watermark.
Confirmation: Full expanded prompt in table.
Niche checklist
- - Prefer one clear metaphor per middle window.
- Punchline must be one screenful—no paragraph.
Workflow
- 1. Text / image / multi-image (SEEDANCE20 only, ≤3 URLs).
- Collect which stats or angles to imply (user may give Chinese—translate lines to English on screen).
- Expand with timed captions + motion graphics language.
- Validate JSON vs frozen tables.
- Confirm →
wait. [Video](url) only for playable links.
CLI reference
~~~sh
node scripts/videogen.js wait --json '{"model":"KLINGV30PRO","prompt":"…","duration":10,"aspectratio":"9:16","generateaudio":true,"negative_prompt":"illegible text, garbled numbers, watermark"}'
node scripts/videogen.js wait --json '{"model":"SEEDANCE20","prompt":"…","duration":10,"aspectratio":"9:16","resolution":"720p","generate_audio":true}'
~~~
Full reference: WERYAI_VIDEO_API.md.
Definition of done
Video or error; prompt includes 0–2 / 2–4 / 4–6 / 6–9 / 9–10 (or user-approved adjustment) subtitle schedule.
Boundaries (out of scope)
- - Not verified census data—illustrative motion; user owns factual claims.
- No
negative_prompt on Seedance; no unsupported fields. resources/WERYAI_VIDEO_API.md for CLI details.
Example prompts
- - INLINECODE21
- INLINECODE22
- INLINECODE23
- INLINECODE24
Model and API constraints (frozen for this skill)
Text-to-video
| modelkey | durations | aspectratios | resolutions | audio | negativeprompt | prompt limit |
|---|
| KLINGV30PRO | 5, 10, 15 | 9:16, 1:1, 16:9 | (omit) | yes | yes | 2000 |
| KLINGV30_STA |
5, 10, 15 | 9:16, 1:1, 16:9 |
(omit) | yes | yes | 2000 |
| SEEDANCE
20 | 5, 10, 15 | 9:16, 1:1, 16:9 | 480p, 720p | yes | no | 2000 |
Image-to-video
| modelkey | durations | aspectratios | resolutions | audio | negativeprompt | image slots |
|---|
| KLINGV30PRO | 5, 10, 15 | 9:16, 16:9, 1:1 | (omit) | yes | yes | single INLINECODE25 |
| KLINGV30_STA |
5, 10, 15 | 9:16, 16:9, 1:1 |
(omit) | yes | yes | single
image |
| SEEDANCE
20 | 5, 10, 15 | 9:16, 1:1, 16:9 | 480p, 720p | yes | no | up to
3 images |
Recommended models
| Tier | modelkey | Default duration |
|---|
| ⭐ Best | KLINGV30PRO | 10 |
| 👍 Good |
KLING
V30_STA | 10 |
| ⚡ Fast | SEEDANCE
20 | 10 (
resolution:
720p) |
Default parameters
| Field | Value |
|---|
| aspect_ratio | 9:16 |
| duration |
10 |
| generate_audio | true |
| resolution | 720p (Seedance) |
| negative_prompt | Kling:
illegible text, garbled captions, watermark |
Text-to-video (primary)
User lists themes → you assign lines + seconds → confirm → wait.
Image-to-video (optional)
Charts / icons stills → animate under same windows.
Tips
- - If user gives too many themes, merge into three middle beats.
- Punchline hits harder with a contrast number (still short).
Packaged as world-as-100-people-video-gen.
世界化为100人(图标+数据竞赛+点睛句)
模板(假设duration为10,除非按比例缩放): 0–2秒 将世界缩小为100人;2–9秒 快速视觉变化(财富分配、性别、水资源/互联网获取等——用户选择主题);9–10秒 一条令人震惊或难忘的数据行(用户主题)。屏幕英文文字配合精确时间窗口。未指定时默认duration为10。
依赖项: WERYAIAPIKEY + Node.js 18+。scripts/videogen.js + WERYAIVIDEOAPI.md。无需其他Cursor技能。
前置条件
- - WERYAIAPIKEY 必须设置;Node 18+。优先使用公开https图片URL;如果整合的scripts/video_gen.js支持本地文件路径,请审查/验证脚本并在本地读取上传至WeryAI前明确同意。
- JSON中model非空。
安全、密钥与API主机
- - WERYAIAPIKEY:机密;切勿提交。
- API主机固定于videogen.js中(与同类技能一致)——仅从环境变量读取密钥。
- 本地图片处理说明:优先使用公开https图片URL。如果整合的scripts/videogen.js支持本地文件路径,它可能读取本地图片并上传至WeryAI以获取公开URL;使用该路径前需审查/验证并获取明确同意。
- 更高保障:优先使用短期或隔离环境;生产环境使用前审查scripts/video_gen.js。验证运行时是否能读取本地图片文件并上传至WeryAI,使用该路径前获取明确同意。
提示词扩展(必选)
- - 时长: 默认10;允许且要求时可为5或15。
- 时间线缩放: 按比例缩放钩子/中间/点睛句段落;保持钩子 → 三个中间节拍 → 最终点睛句顺序。
锁定节奏(以时长10为例):
- - 0.0–2.0秒 — 钩子:例如将地球缩小为100人。
- 2.0–4.0秒、4.0–6.0秒、6.0–9.0秒 — 三条轮换的数据/主题行(用户简报):例如财富转移 / 性别 / 谁得到什么——保持标签简短;每条搭配图标变形、饼图擦除或计数器跳动。
- 9.0–10.0秒 — 点睛数据(用户意图的一条简短英文行——例如仅有X拥有Y)。
视觉效果: 等距或扁平UI图标、发光、利落转场(约1秒);无微小不可读数字,除非字幕承载可读数据。
排版: 粗体无衬线、轮廓、下三分之一;仅英文,除非用户另有要求。
音频: 若开启音频,轻量数字滴答声/嗖嗖声。
负面提示(Kling): 乱码数字、难以辨认的文字、水印。
确认: 表格中完整的扩展prompt。
细分检查清单
- - 每个中间窗口优先使用一个清晰隐喻。
- 点睛句必须一屏显示——无段落。
工作流程
- 1. 文本/图片/多图(仅SEEDANCE20,≤3个URL)。
- 收集需暗示的数据或角度(用户可能提供中文——将行翻译为屏幕英文)。
- 使用定时字幕+动态图形语言进行扩展。
- 验证JSON与固定表格。
- 确认 → wait。
- 仅返回Video用于可播放链接。
CLI参考
~~~sh
node scripts/videogen.js wait --json {model:KLINGV30PRO,prompt:…,duration:10,aspectratio:9:16,generateaudio:true,negative_prompt:illegible text, garbled numbers, watermark}
node scripts/videogen.js wait --json {model:SEEDANCE20,prompt:…,duration:10,aspectratio:9:16,resolution:720p,generate_audio:true}
~~~
完整参考: WERYAIVIDEO_API.md。
完成标准
返回Video或错误;prompt包含0–2 / 2–4 / 4–6 / 6–9 / 9–10(或用户批准的调整)字幕时间表。
边界(范围外)
- - 非验证的人口普查数据——示意性动态;用户对事实主张负责。
- Seedance无negativeprompt;无不支持字段。
- CLI详情见resources/WERYAIVIDEO_API.md。
示例提示词
- - 主题:互联网接入、教育、财富——点睛句关于不平等
- 气候+水+能源图标,最后一行关于可再生能源占比
- 三个图表图片URL → Seedance多图,相同字幕窗口
- 医疗保健获取与财富百分位对比——点睛句关于谁被排除在外
模型与API约束(本技能固定)
文生视频
| 模型键 | 时长 | 宽高比 | 分辨率 | 音频 | 负面提示 | 提示词限制 |
|---|
| KLINGV30PRO | 5, 10, 15 | 9:16, 1:1, 16:9 | (省略) | 是 | 是 | 2000 |
| KLINGV30STA |
5, 10, 15 | 9:16, 1:1, 16:9 |
(省略) | 是 | 是 | 2000 |
| SEEDANCE
20 | 5, 10, 15 | 9:16, 1:1, 16:9 | 480p, 720p | 是 | 否 | 2000 |
图生视频
| 模型键 | 时长 | 宽高比 | 分辨率 | 音频 | 负面提示 | 图片槽位 |
|---|
| KLINGV30PRO | 5, 10, 15 | 9:16, 16:9, 1:1 | (省略) | 是 | 是 | 单张image |
| KLINGV30STA |
5, 10, 15 | 9:16, 16:9, 1:1 |
(省略) | 是 | 是 | 单张image |
| SEEDANCE
20 | 5, 10, 15 | 9:16, 1:1, 16:9 | 480p, 720p | 是 | 否 | 最多
3张images |
推荐模型
| 等级 | 模型键 | 默认时长 |
|---|
| ⭐ 最佳 | KLINGV30PRO | 10 |
| 👍 良好 |
KLINGV3
0STA | 10 |
| ⚡ 快速 | SEEDANCE
20 | 10(resolution:
720p) |
默认参数
| 字段 | 值 |
|---|
| aspect_ratio | 9:16 |
| duration |
10 |
| generate_audio | true |
| resolution | 720p(Seedance) |
| negative_prompt | Kling:illegible text, garbled captions, watermark |
文生视频(主要)
用户列出主题 → 你分配行+秒数 → 确认 → wait。
图生视频(可选)
图表/图标静态图 → 在相同窗口下进行动画处理。
提示
- - 如果用户给出过多主题,合并为三个中间节拍。
- 点睛句搭配对比数字效果更佳(仍需简短)。
打包为world-as-100-people-video-gen。