Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
t0ugh avatar

Video Init

  • 31 installs
  • 110 repo stars
  • Updated May 18, 2026
  • t0ugh/videoclaw

Helps with ai & agent building tasks.

About

video-init is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted coding.

  • video-init
  • AI & Agent Building
  • AI-coding skill

Video Init by the numbers

  • 31 all-time installs (skills.sh)
  • +4 installs in the week ending Aug 5, 2026 (Skillselion tracking)
  • Ranked #9,202 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
  • Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/t0ugh/videoclaw --skill video-init

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs31
repo stars110
Last updatedMay 18, 2026
Repositoryt0ugh/videoclaw

What it does

Helps with ai & agent building tasks.

Files

SKILL.mdMarkdownGitHub ↗

video-init - 初始化项目

概述

初始化新的视频项目,创建目录结构。交互确认在 skill 层通过 AskUserQuestion 实现

触发条件

用户说"创建一个新项目"、"初始化项目"或类似表达时触发。

交互流程

Step 1: 询问项目名称

用 AskUserQuestion 询问:

"请给项目起个名字?"

Step 2: 询问图像生成提供商

选项:
1) volcengine (火山引擎 Seedream)
2) dashscope (阿里云)
3) gemini (Google)
4) mock (测试用)

Step 3: 询问视频生成提供商

选项:
1) volcengine (火山引擎 Seedance)
2) dashscope (阿里云 Wan2.6)
3) mock (测试用)

Step 4: 询问存储方式

选项:
1) local (本地存储)
2) google_drive (上传到 Google Drive)

执行流程

1. 用 AskUserQuestion 依次收集配置(项目名称、图像提供商、视频提供商、存储方式) 2. 调用 videoclaw init <project-name> 创建项目 3. 用 videoclaw config --project <name> --set 设置项目配置(推荐使用新的 backend 字段;如果已有 Codex / ChatGPT 登录环境,建议优先用 Codex-first 路径)

# 示例执行
videoclaw init my-video
videoclaw config --project my-video --set models.image.backend=codex-host-image
videoclaw config --project my-video --set models.image.model=gpt-image-2-medium
videoclaw config --project my-video --set models.image.auth=chatgpt_login
videoclaw config --project my-video --set models.image.transport=codex_oauth
videoclaw config --project my-video --set models.video.provider=volcengine
videoclaw config --project my-video --set storage.provider=google_drive

创建的目录

  • assets/ - 共享角色、场景、道具、封面
  • videos/ - 多个视频单元(每个 video 独立管理)
  • exports/ - 最终交付产物

参数

  • project_name: 项目名称(必填)
  • --dir: 项目目录路径(可选,默认 ~/videoclaw-projects)

全局配置(可选)

用户也可以先设置全局配置,新项目会自动使用:

# 设置全局默认提供商
videoclaw config --set models.image.backend=volcengine
videoclaw config --set models.video.provider=volcengine
videoclaw config --set storage.provider=google_drive

示例

用户: 帮我创建一个新项目

Claude Code:
  - 询问项目名称
  - 询问图像提供商(volcengine/dashscope/gemini/mock)
  - 询问视频提供商(volcengine/dashscope/mock)
  - 询问存储方式(local/google_drive)
  - 执行: videoclaw init my-video
  - 执行: videoclaw config --project my-video --set models.image.backend=xxx
  - 执行: videoclaw config --project my-video --set models.video.provider=xxx
  - 执行: videoclaw config --project my-video --set storage.provider=xxx

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.