AI Video Script Generator (Yijia Test)
生成 AI 短视频脚本、镜头拆解、画面提示词、字幕文案与可直接进入视频生成测试的 prompt。用于把一个主题快速整理成可执行的视频方案,尤其适合:短视频选题、分镜脚本、AI 视频提示词、Yijia/Sora/Grok 文生视频测试、图生视频测试、爆款短视频结构化策划。
MIT-0 · Free to use, modify, and redistribute. No attribution required.
⭐ 0 · 85 · 0 current installs · 0 all-time installs
MIT-0
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (generate short-video scripts, shot breakdowns, and generation prompts) matches the files and declared requirements: no binaries, no env vars, no installs — everything is instruction-only and focused on prompt/script generation.
Instruction Scope
SKILL.md stays on-task (templates, examples, prompt compression). The included reference file contains example local test commands (python/bash using /Users/shift/...) and mentions Yijia API keys for testing; those are informational and not invoked by the skill itself, but if a user follows them they will run local scripts and may require credentials.
Install Mechanism
No install spec and no code files — the skill will not write or execute code on disk during install. Low installation risk.
Credentials
The skill declares no required environment variables or credentials. The reference docs warn that running the creator/test workflow may require a Yijia executable or API key — this is relevant only if the user chooses to execute those local test commands.
Persistence & Privilege
always:false and no special privileges requested. Agent-autonomous invocation is allowed (platform default) but the skill itself does not request persistent or elevated access.
Assessment
This skill appears internally consistent and low-risk: it only generates script text and prompts and does not request credentials or install software. Two practical cautions: (1) the bundled reference shows example shell commands that run local scripts and note that a Yijia API key may be needed — do not run those commands or paste secrets unless you trust the source and have inspected the scripts. (2) The skill's source is unknown; if you plan to integrate the generated prompts into automated workflows or to run the referenced creator scripts, review those local scripts and supply API keys only to trusted tools. Otherwise it is safe to use for manual prompt/script generation.Like a lobster shell, security has layers — review code before you run it.
Current versionv1.1.0
Download ziplatest
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
SKILL.md
AI Video Script Generator
把一个模糊主题整理成可执行的视频脚本包:
- 标题与定位
- 时长建议
- 分镜表
- 逐镜头提示词
- 旁白/字幕文案
- 可直接用于 Yijia 测试的视频总 prompt
输出目标
默认输出 5 个部分:
- 视频概述
- 标题
- 目标受众
- 风格
- 建议时长
- 分镜表
- 时间段
- 画面
- 镜头运动
- 情绪/节奏
- 逐镜头提示词
- 画面主体
- 场景环境
- 光线与镜头语言
- 风格关键词
- 文案层
- 开头钩子
- 旁白
- 字幕
- 结尾 CTA
- 生成测试 prompt
- 一段适合直接送入 Yijia/Sora/Grok 的整合版 prompt
推荐工作流
场景 A:先做脚本
当用户只说主题时,按下面顺序输出:
- 先定义视频定位
- 再给 3-5 个分镜
- 再给每镜头提示词
- 再给字幕与旁白
- 最后给一段可直接生成视频的主 prompt
场景 B:直接为 AI 视频生成服务
如果目标是直接测试文生视频/图生视频:
- 优先输出 8秒 / 10秒 / 15秒 版本
- 画面描述要具体
- 避免镜头过多
- 强调主体、动作、场景、镜头、光线、风格、画质
- 结尾给一个压缩后的单段 prompt,便于直接粘贴到视频生成器
场景 C:适配短视频平台
短视频优先结构:
- 0-2 秒:钩子
- 2-6 秒:核心动作/反差
- 6-8 秒或 6-10 秒:结果/反转/记忆点
适配 Yijia 的写法
当目标是 Yijia 测试时,优先遵循:
- 单视频只保留一个核心主体
- 动作描述清晰,不要多事件并发
- 场景尽量单一
- 镜头运动只选 1-2 个
- 风格关键词不超过 6 个
- 避免过长抽象说明
推荐总 prompt 结构:
主体 + 动作 + 场景 + 镜头运动 + 光线/氛围 + 风格 + 画质
示例:
一只橘猫坐在雨后窗边慢慢转头看向镜头,桌上有一杯冒着热气的咖啡,镜头缓慢推进,室内暖色傍晚光线,治愈、细腻、电影感,竖屏,高清
建议输出模板
# 标题
## 概述
- 时长:8秒 / 10秒 / 15秒
- 风格:
- 受众:
- 目标:
## 分镜
### 镜头1
- 时间:0-2秒
- 画面:
- 镜头:
- 提示词:
- 配音/字幕:
### 镜头2
...
## 完整旁白
## 字幕版文案
## Yijia 测试总 Prompt
## 可选变体
- 轻松版
- 电影感版
- 爆款钩子版
快速模板
1) 萌宠治愈类
主体可爱、动作简单、场景温暖、镜头缓慢、光线柔和、情绪治愈
2) 搞笑反转类
前半段正常,后半段出现反差动作,结尾有记忆点,节奏快,画面明确
3) 解压沉浸类
主体重复性动作,特写,慢节奏,细节质感强,ASMR 氛围明显
参考
若要把脚本直接接到 creator 测试流程,读取:
references/yijia-test-workflow.md
Files
4 totalSelect a file
Select a file to preview.
Comments
Loading comments…
