← Back to Skills Marketplace
infiniteask

生成短视频分镜脚本_无限

by 未来X无限 · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
129
Downloads
1
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install generate-short-video-storyboard-script-infinite
Description
自动化生成短视频分镜脚本。打开Chrome浏览器访问DeepSeek平台,使用深度思考和智能搜索模式,根据旁白文本自动生成分镜脚本。触发场景:生成分镜脚本、生成短视频分镜、分镜脚本生成、旁白转分镜、短视频分镜制作。
README (SKILL.md)

生成短视频分镜脚本

自动化流程:使用 DeepSeek 平台将旁白文本转换为分镜脚本。

触发场景

用户说:

  • "生成分镜脚本"
  • "生成短视频分镜脚本"
  • "生成视频分镜脚本"

浏览器配置

执行步骤

1. 启动浏览器

  1. 调用 browser(action=start, target="host", profile="openclaw") 启动浏览器
  2. 若浏览器已在运行,跳过此步骤
  3. 检查 browser(action=status, target="host") 确认状态
  4. 导航到 DeepSeek 平台:https://chat.deepseek.com。

2. 开启新会话

在 DeepSeek 页面左上角点击「开启新会话」按钮。

3. 配置生成模式

在右侧框架中,点击专家模式,同时开启「深度思考」和「智能搜索」两个模式。

重要说明

  • 两个按钮是独立的,可以同时开启

操作步骤

  1. 两个模式都开启后,继续下一步

4. 输入分镜提示词

在提示词输入框中输入以下固定模板:

我要用后面的旁白做一个短视频。你把它拆成很多个分镜,每一个分镜只有画面描述和旁白。画面描述内容前直接是"分镜"加分镜的序号加":"(无需输出“画面描述:”这5个字符。),每一个分镜只配少量的旁白(“旁白:”三个字符需输出。每个分镜的旁白无需用双引号括起来)。不要增删改旁白。分镜总字符数小于3000个。

5. 追加旁白内容

在分镜提示词后面,用双引号括起旁白内容追加到输入框。

格式示例

我要用后面的旁白做一个短视频。你把它拆成很多个分镜,每一个分镜只有画面描述和旁白。画面描述内容前直接是"分镜"加分镜的序号加":"(无需输出“画面描述:”这5个字符。),每一个分镜只配少量的旁白(“旁白:”三个字符需输出。每个分镜的旁白无需用双引号括起来)。不要增删改旁白。分镜总字符数小于3000个。"这里是用户的旁白内容..."

6. 提交对话

点击提示词输入框右下角的向上箭头按钮提交对话。

输出格式

DeepSeek 将返回分镜脚本,格式如下:

分镜1: [画面描述]
旁白:[旁白片段]

分镜2: [画面描述]
旁白:[旁白片段]

...

使用方式

调用此技能时,用户需提供:

  1. 旁白文本内容

示例调用:

帮我生成分镜脚本,旁白是:"今天我们来聊聊人工智能的发展历程。从最初的图灵测试,到如今的ChatGPT,AI已经走过了漫长的道路..."
Usage Guidance
This skill appears to do what it claims: it automates entering your narration into the DeepSeek web UI to generate storyboards. Before installing or using it, consider: (1) any narration you supply will be sent to a third-party site (chat.deepseek.com) — avoid pasting secrets, personal data, or proprietary information; (2) the skill asks to use a named Chrome profile (openclaw) — if that profile contains logins or cookies, those could be accessible in the session, so prefer an ephemeral/incognito profile or a profile without sensitive accounts; (3) check DeepSeek's privacy/terms if you care about retention or reuse of submitted text. Other than these privacy considerations, the skill is coherent and does not request disproportionate access or installs.
Capability Analysis
Type: OpenClaw Skill Name: generate-short-video-storyboard-script-infinite Version: 1.0.1 The skill is a legitimate automation tool designed to generate video storyboard scripts by interacting with the DeepSeek AI platform. It uses standard browser automation commands (SKILL.md) to navigate to chat.deepseek.com, configure specific search modes, and format user-provided text into a storyboard structure. No indicators of data exfiltration, malicious execution, or harmful prompt injection were found.
Capability Assessment
Purpose & Capability
Name/description match the instructions: the SKILL.md directs the agent to open Chrome, navigate to chat.deepseek.com, enable specific modes, paste the provided narration, and submit — exactly what a DeepSeek-driven storyboard generator would need. No unrelated binaries, env vars, or installs are requested.
Instruction Scope
Instructions are narrowly scoped to controlling a browser session on DeepSeek and formatting the prompt. However, the runtime flow sends the user's narration to an external site (chat.deepseek.com) and asks the agent to reuse a named browser profile ('openclaw'), which could expose cookies/session state. This is a privacy/ data-exfiltration risk depending on the sensitivity of the narration and the profile contents.
Install Mechanism
No install spec or code is included (instruction-only). Nothing is downloaded or written to disk by the skill itself.
Credentials
The skill requires no environment variables, credentials, or config paths. That is proportionate to its described purpose. Note: it implicitly relies on the agent having a controllable Chrome browser and the 'openclaw' profile, which are platform capabilities rather than secrets.
Persistence & Privilege
The skill does not request always:true and is user-invocable only. It instructs using a named browser profile; reusing an existing profile could expose other session data. There is no evidence it modifies other skills or system settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install generate-short-video-storyboard-script-infinite
  3. After installation, invoke the skill by name or use /generate-short-video-storyboard-script-infinite
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Initial release. - Automates the generation of short video storyboard scripts from provided narration. - Integrates with DeepSeek via Chrome, using both "Deep Thinking" and "Smart Search" modes. - Guides users step-by-step from browser launch to script output. - Accepts narration as input and outputs segmented storyboard with scene descriptions and matching narration.
v1.0.0
- Initial release: Automates generation of short video storyboard scripts from provided narration text. - Integrates with DeepSeek platform via Chrome for intelligent, multi-mode script generation. - Supports automatic browser launching and session configuration. - Accepts user narration and formats outputs as numbered storyboard frames with concise pairing of narration and visual description. - Output constrained to under 3000 Chinese characters for efficiency and readability.
Metadata
Slug generate-short-video-storyboard-script-infinite
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is 生成短视频分镜脚本_无限?

自动化生成短视频分镜脚本。打开Chrome浏览器访问DeepSeek平台,使用深度思考和智能搜索模式,根据旁白文本自动生成分镜脚本。触发场景:生成分镜脚本、生成短视频分镜、分镜脚本生成、旁白转分镜、短视频分镜制作。 It is an AI Agent Skill for Claude Code / OpenClaw, with 129 downloads so far.

How do I install 生成短视频分镜脚本_无限?

Run "/install generate-short-video-storyboard-script-infinite" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 生成短视频分镜脚本_无限 free?

Yes, 生成短视频分镜脚本_无限 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 生成短视频分镜脚本_无限 support?

生成短视频分镜脚本_无限 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 生成短视频分镜脚本_无限?

It is built and maintained by 未来X无限 (@infiniteask); the current version is v1.0.1.

💬 Comments