← Back to Skills Marketplace
2719040953

Jianying Video Compose

by 2719040953 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
432
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install jianying-video-compose
Description
剪映API视频合成自动化。通过剪映代理API完成视频全流程制作,包括草稿创建、素材添加(图片/视频/音频)、文本字幕编辑、特效处理、云渲染导出。适用于需要批量生成视频、自动合成短视频、动态字幕视频等场景。
README (SKILL.md)

剪映API视频合成

快速开始

0. 上传素材(可选)

如果素材不在OSS上,先上传:

POST https://np-newsmgr-uat.eastmoney.com/video-admin/api/oss/upload
Content-Type: multipart/form-data

参数:file(文件)、bucket(桶名称)、filename(文件名) 返回:ossKey(可拼接素材地址)

1. 创建草稿

POST https://np-newsmgr-uat.eastmoney.com/videomake/api/capcut-proxy/create_draft
{"width": 1080, "height": 1920}

返回 draft_id

2. 添加素材

  • 图片: add_image - 背景图、产品图
  • 视频: add_video - 视频片段
  • 音频: add_audio - 背景音乐、音效
  • 文本: add_text / add_subtitle - 标题、字幕

3. 导出视频

POST https://np-newsmgr-uat.eastmoney.com/videomake/api/capcut-proxy/export_draft
{"draft_id": "xxx", "resolution": "1080P", "framerate": 30}

⚠️ 注意: 导出视频是云渲染过程,平均耗时约 80秒,请耐心等待返回的 video_file_url

核心概念

relative_index 图层顺序

  • 值越小越在下层,值越大越在上层
  • 背景图用 relative_index: 0(最底层)
  • 文字/图片/视频用 relative_index: 1,2,3...(在上层)

track_name 轨道概念

  • 相同时间段内,同一轨道(相同track_name)后者覆盖前者
  • 不同轨道可叠加显示,按relative_index决定上下层

素材地址拼接

http://np-newsmgr-uat.emapd.com/videomake/api/resource/download/bucket?filename={ossKey}&bucket={bucket}
  • 默认bucket: np-vediooss-material

常用配置

字幕默认配置

  • 字体: 抖音美好体
  • 字号: 10
  • 位置: transform_y: -0.42(底部居中)
  • 颜色: #FFFFFF
  • 入场动画: 放大,0.5秒
  • 出场动画: 缩小,0.5秒

可用字体

抖音美好体、思源黑体、OPPO Sans、挥墨体、文轩体、鸿蒙OS、小米MiSans、站酷快乐体

文本动画

放大、缩小、渐显、溶解、打字机、弹性伸缩、弹簧、闪光、旋转飞入、模糊发光、逐字弹跳

详细参数

完整API参数说明见 references/api_reference.md

Usage Guidance
This skill appears to be a working instruction set for automating video composition, but exercise caution before installing or using it. Key considerations: - The documented API hosts are UAT/test domains (np-newsmgr-uat.eastmoney.com and emapd.com), not an official Jianying/CapCut domain. Confirm who operates these endpoints and whether they are intended for production use. - The API is documented as unauthenticated (no API key). Do NOT upload sensitive or private media to these endpoints unless you trust the operator and have explicit data-retention guarantees. - If you need production-grade integration, ask the skill author for the official API endpoint, authentication details, and a privacy/data-retention policy. - When testing, use only non-sensitive sample media and monitor network activity. Consider sandboxing or restricting the agent's ability to upload files until you verify the service. What would change this assessment to 'benign': explicit, authoritative documentation or homepage showing these endpoints are official/properly managed (not UAT), and clear authentication/data-handling guidance from the owner. If you cannot verify ownership and intent, treat the skill as untrusted for sensitive data.
Capability Analysis
Type: OpenClaw Skill Name: jianying-video-compose Version: 1.0.0 The skill bundle provides instructions and API documentation for an AI agent to automate video composition using a CapCut proxy service hosted on East Money's UAT environment (eastmoney.com). It includes standard video editing capabilities such as uploading assets, managing drafts, adding text/media, and triggering cloud rendering. While the documentation in SKILL.md and references/api_reference.md indicates that the target API endpoints currently lack authentication, this is a characteristic of the external service rather than a malicious feature of the skill itself. The skill's logic is entirely aligned with its stated purpose of video automation and contains no evidence of data exfiltration, malicious execution, or prompt injection.
Capability Assessment
Purpose & Capability
The skill claims to automate video composition via a '剪映代理API' and the SKILL.md provides concrete REST endpoints and parameters that implement that functionality (create draft, add_image/add_video/add_text, export_draft). However the API base URLs are on np-newsmgr-uat.eastmoney.com and emapd.com (UAT/test hostnames), not an official Jianying/CapCut domain. The use of a third-party UAT proxy is unexpected and not explained in the metadata, which reduces confidence that the endpoints align with the claimed production service.
Instruction Scope
The SKILL.md explicitly instructs uploading files (multipart/form-data) and sending media (image/video/audio) to external endpoints and then composing/exporting via the proxy API. These instructions do not request local system secrets, but they direct potentially sensitive file uploads to an external, unauthenticated host. The docs state "无需apikey" (no API key) which means calls to these endpoints may be unauthenticated and therefore could be logged or retained by the remote service. The skill gives the agent broad discretion to upload and reference files via URLs; that behavior could leak sensitive content if the user or agent supplies private files.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to write to disk. That's the lowest install risk: nothing is installed or executed locally by the skill itself.
Credentials
The skill requests no environment variables or credentials, which is consistent with the SKILL.md claim that the API requires no API key. However, the absence of authentication increases the risk that uploading media or draft data will be exposed to the proxy operator. Also the skill provides default OSS bucket and resource download URLs; these defaults mean the skill will interact with remote storage without credential gating, which may be inappropriate for sensitive content.
Persistence & Privilege
The skill does not request persistent presence (always: false) and does not modify agent configuration. Autonomous invocation is allowed by default but is not an additional privilege requested by the skill beyond platform defaults.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install jianying-video-compose
  3. After installation, invoke the skill by name or use /jianying-video-compose
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of jianying-video-compose: - Automates video creation via Jianying (Capcut) proxy API. - Supports full workflow: draft creation, adding images, video, audio, and subtitles, effects, cloud rendering, and export. - Batch video generation and automated short video synthesis enabled. - Provides config details for tracks, layers, material upload, and subtitle presets. - Includes quickstart guide and core API usage examples.
Metadata
Slug jianying-video-compose
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Jianying Video Compose?

剪映API视频合成自动化。通过剪映代理API完成视频全流程制作,包括草稿创建、素材添加(图片/视频/音频)、文本字幕编辑、特效处理、云渲染导出。适用于需要批量生成视频、自动合成短视频、动态字幕视频等场景。 It is an AI Agent Skill for Claude Code / OpenClaw, with 432 downloads so far.

How do I install Jianying Video Compose?

Run "/install jianying-video-compose" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Jianying Video Compose free?

Yes, Jianying Video Compose is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Jianying Video Compose support?

Jianying Video Compose is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Jianying Video Compose?

It is built and maintained by 2719040953 (@2719040953); the current version is v1.0.0.

💬 Comments