← 返回 Skills 市场
Json To Jianying Description
作者
2719040953
· GitHub ↗
· v1.0.0
· MIT-0
103
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install json-to-jianying-description
功能描述
将特定的JSON视频素材格式(包含oralBroadcastingList、materialList、bgmInfo等字段)转换为剪映API可用的自然语言视频描述。用于根据素材JSON生成视频制作指令,包括素材匹配、时间轴计算、字幕时间同步、特殊效果处理等。当用户发送包含oralBroadcastingList和...
使用说明 (SKILL.md)
JSON 转剪映视频描述
输入格式
用户会发送一个包含以下关键字段的 JSON:
{
"oralBroadcastingList": [
{
"oralText": "文案内容",
"materialDesc": "素材描述",
"materialKey": "desc_material_xxx.jpg",
"ttsResult": {
"text": "实际朗读文本",
"duration": 2.28,
"audioOssBucket": "np-vediooss-material",
"audioOssKey": "volc_tts_xxx.mp3",
"captionStartEnds": [
{ "caption": "字幕1", "startSecond": 0.0, "endSecond": 1.12, "duration": 1.12 },
{ "caption": "字幕2", "startSecond": 1.12, "endSecond": 2.28, "duration": 1.16 }
]
}
}
],
"materialList": [
{
"ossBucket": "np-vediooss-material",
"ossKey": "desc_material_xxx.jpg",
"url": "https://...",
"desc": "素材描述"
}
],
"bgmInfo": {
"ossBucket": "np-vediooss-material",
"ossKey": "desc_material_xxx.mp3",
"url": "https://...",
"title": "背景音乐标题",
"duration": 193000
}
}
输出格式
转换为自然语言描述,格式如下:
视频总时长:X秒
0-X秒是 {materialKey} 这个素材,对应的字幕:
- 0.0-1.21 秒是 "字幕1"
- 1.21-1.91 秒是 "字幕2"
...
X-Y秒是 {materialKey} 这个素材,对应的字幕:
...
背景图:{ossKey}
特殊需求:
- 在某某时间段插入图片:{ossKey}
- 给某素材添加某某特效
转换步骤
1. 解析 oralBroadcastingList
- 遍历
oralBroadcastingList数组 - 每个元素代表一个视频片段
- 使用
ttsResult.captionStartEnds计算该片段的起止时间
2. 计算时间轴
- 起始时间:累加前面所有片段的 duration
- 结束时间:起始时间 + 当前片段的 ttsResult.duration
- 每个字幕的绝对时间 = 片段起始时间 + captionStartEnds 中的相对时间
3. 匹配素材
- 从
materialList中查找ossKey与oralBroadcastingList[].materialKey匹配的素材 - 使用
url字段作为素材地址
4. 处理背景图
- 默认使用
materialList中的第一张图片作为背景图 - 或使用用户指定的背景图
5. 处理 BGM
- 提取
bgmInfo.url作为背景音乐
6. 特殊需求识别
用户可能会指定额外需求,如:
- 在某时间段插入额外图片
- 给某素材添加特效(如"放映滚动")
- 调整素材顺序
素材地址拼接
图片/视频/音频素材地址拼接:
- 基础URL:
http://np-newsmgr-uat.emapd.com/videomake/api/resource/download/bucket - 拼接方式:
${baseUrl}?filename=${ossKey}&bucket=${ossBucket}
示例:
- ossKey:
深色质感背景图3.png,bucket:np-vediooss-material - 完整地址:
http://np-newsmgr-uat.emapd.com/videomake/api/resource/download/bucket?filename=深色质感背景图3.png&bucket=np-vediooss-material
完整示例
输入 JSON(用户提供的例子)
{
"oralBroadcastingList": [
{
"oralText": "宝济药业 - B有重大进展!",
"materialKey": "desc_material_D36841A87F2909BB35DC950C500064D9.jpg",
"ttsResult": {
"captionStartEnds": [
{ "caption": "宝济药业 - B", "startSecond": 0.0, "endSecond": 1.12 },
{ "caption": "有重大进展", "startSecond": 1.12, "endSecond": 2.28 }
],
"duration": 2.28
}
},
{
"oralText": "3月25日,宝济药业 - B公告相关进展。",
"materialKey": "desc_material_D36841A87F2909BB35DC950C500064D9.jpg",
"ttsResult": {
"captionStartEnds": [
{ "caption": "3月25日", "startSecond": 0.0, "endSecond": 1.03 },
{ "caption": "宝济药业 - B", "startSecond": 1.03, "endSecond": 1.92 },
{ "caption": "公告相关进展", "startSecond": 1.92, "endSecond": 3.14 }
],
"duration": 3.14
}
}
// ... more items
],
"materialList": [
{
"ossKey": "desc_material_D36841A87F2909BB35DC950C500064D9.jpg",
"ossBucket": "np-vediooss-material",
"url": "https://img0.baidu.com/...",
"desc": "宝济药业大楼外观局部"
}
],
"bgmInfo": {
"ossBucket": "np-vediooss-material",
"ossKey": "desc_material_C6CF433C3F6ADEBA88682E05EE9D9F5F.mp3",
"url": "https://cdnsaas.kuai.360.cn/...",
"title": "氛围音乐 助眠"
}
}
输出描述
视频总时长:29.39秒
0-2.28秒是 desc_material_D36841A87F2909BB35DC950C500064D9.jpg 这个素材,对应的字幕:
- 0.0-1.12 秒是 "宝济药业 - B"
- 1.12-2.28 秒是 "有重大进展"
2.28-5.42秒是 desc_material_D36841A87F2909BB35DC950C500064D9.jpg 这个素材,对应的字幕:
- 2.28-3.31 秒是 "3月25日"
- 3.31-4.20 秒是 "宝济药业 - B"
- 4.20-5.42 秒是 "公告相关进展"
... (以此类推)
背景图:desc_material_D36841A87F2909BB35DC950C500064D9.jpg
BGM:氛围音乐 助眠 (url: https://cdnsaas.kuai.360.cn/...)
注意事项
- 时间累加:每个片段的绝对起始时间需要累加前面所有片段的 duration
- 素材匹配:materialKey 可能不完整(如只有文件名),需要在 materialList 中模糊匹配
- OSS 地址拼接:需要将 ossKey 和 ossBucket 拼接成完整的素材地址
- 用户额外需求:用户可能会指定插入图片、添加特效等额外需求,需要在描述中体现
使用场景
当用户发送类似结构的 JSON 并要求:
- "帮我生成一个视频"
- "把这个转成视频描述"
- "用这个素材做视频"
此时应使用此 skill 进行 JSON 解析和描述转换。
安全使用建议
This skill is internally consistent and asks for nothing sensitive. Before using it in production: 1) review the hard-coded base URL (np-newsmgr-uat.emapd.com) — it looks like a UAT/internal endpoint; confirm you want generated descriptions to reference that domain or replace it with your production asset host; 2) be cautious if you (or an agent calling this skill) plan to fetch asset URLs returned by the skill — ensure those URLs point to resources you control and do not expose private material; 3) because this is instruction-only, the skill won't install code, but validate any downstream automation that consumes the generated descriptions (e.g., an automated fetch/upload step) so it doesn't inadvertently transmit sensitive data.
功能分析
Type: OpenClaw Skill
Name: json-to-jianying-description
Version: 1.0.0
The skill is a data transformation utility designed to convert structured JSON video metadata into natural language descriptions for the Jianying (CapCut) API. It performs timeline calculations and string concatenation to generate resource URLs (e.g., at np-newsmgr-uat.emapd.com) based on user-provided OSS keys. No evidence of data exfiltration, malicious execution, or prompt injection was found in SKILL.md or references/json-schema.md.
能力评估
Purpose & Capability
The name/description (convert oralBroadcastingList/materialList/bgmInfo JSON into Jianying/剪映-style video descriptions) matches the SKILL.md's parsing, timeline calculation, material matching and output formatting. No unrelated binaries or environment variables are requested.
Instruction Scope
The instructions are narrowly scoped to parsing the supplied JSON, computing absolute times, matching materialList entries, and producing human-readable instructions. One note: the docs include a hard-coded base URL (http://np-newsmgr-uat.emapd.com/...) used to build asset download links; the skill does not itself instruct fetching those URLs or accessing local files, but consumers should be aware the outputs reference that endpoint.
Install Mechanism
No install spec and no code files — this is instruction-only, so nothing is written to disk or installed.
Credentials
No environment variables, credentials, or config paths are required. The data used comes entirely from user-supplied JSON and the documented base URL/urls inside examples.
Persistence & Privilege
always:false and no special privileges requested. The skill does not modify system/agent settings or other skills.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install json-to-jianying-description - 安装完成后,直接呼叫该 Skill 的名称或使用
/json-to-jianying-description触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
- Initial release of json-to-jianying-description.
- Converts specific JSON video material formats (with oralBroadcastingList, materialList, bgmInfo) to natural language video descriptions compatible with Jianying API.
- Supports time axis calculation, subtitle synchronization, material matching, special effect handling, and address construction.
- Handles user requirements for video production instructions based on input JSON.
- Recognizes and incorporates extra user requests such as inserting images or adding effects.
元数据
常见问题
Json To Jianying Description 是什么?
将特定的JSON视频素材格式(包含oralBroadcastingList、materialList、bgmInfo等字段)转换为剪映API可用的自然语言视频描述。用于根据素材JSON生成视频制作指令,包括素材匹配、时间轴计算、字幕时间同步、特殊效果处理等。当用户发送包含oralBroadcastingList和... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 103 次。
如何安装 Json To Jianying Description?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install json-to-jianying-description」即可一键安装,无需额外配置。
Json To Jianying Description 是免费的吗?
是的,Json To Jianying Description 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Json To Jianying Description 支持哪些平台?
Json To Jianying Description 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Json To Jianying Description?
由 2719040953(@2719040953)开发并维护,当前版本 v1.0.0。
推荐 Skills