douyin-downloader-js

使用 Node.js 从抖音网页解析视频信息并下载，支持视频和图文类型

Security Scan

The skill's code and instructions are consistent with its stated purpose (parsing Douyin pages and downloading media); no unrelated credentials or install steps are requested, but it performs network fetches, follows redirects, and writes files so you should only run it with trusted links or in a sandboxed environment.

Purpose & Capability

Name/description, SKILL.md usage, and the included Node.js script align: the script fetches Douyin pages, extracts window._ROUTER_DATA, parses video/image metadata, and supports downloading media. There are no extraneous environment variables, binaries, or unrelated dependencies declared.

Instruction Scope

The SKILL.md instructs running the bundled Node script with a Douyin link and optional output directory — this matches the code. The script performs outbound HTTP(S) requests, follows redirects, and writes downloaded media to disk (workspace). This is expected for a downloader but means untrusted input (arbitrary URLs) could cause the agent to fetch arbitrary endpoints (SSRF risk) or download unexpected content. The SKILL.md does not explicitly restrict input validation or warn about running on networked servers.

Install Mechanism

No install spec — instruction-only with a bundled script. Nothing is downloaded or executed at install time, which is proportionate for this skill.

Credentials

The skill requires no environment variables, credentials, or special config paths. That is appropriate for a web-scraping/downloading utility and matches the SKILL.md and code.

Persistence & Privilege

The skill is not force-included (always: false) and does not request elevated or persistent platform privileges. It can be invoked by the agent (normal default); nothing in the package attempts to modify other skills or global agent configuration.

Assessment

This skill appears to do what it says: it fetches Douyin pages, parses embedded JSON, and downloads media to disk. Before installing or running it: 1) Only provide trusted Douyin links — the script will follow redirects and can fetch arbitrary HTTP(S) URLs (SSRF risk if run from a server with access to internal networks). 2) Run it in a sandbox or isolated environment (local machine or container) if you care about safety. 3) Review the full script (the CLI/IO portions) yourself to confirm where files are saved and how temporary files are handled. 4) Be mindful of copyright and site terms when downloading content. If you plan to let the agent call this autonomously, restrict the inputs it may pass to avoid unintended network fetches.

抖音视频下载器 (Node.js 版本)

描述

使用 Node.js 从抖音网页的 window._ROUTER_DATA 中解析视频信息，支持：

• 视频类型：下载视频并去除水印
• 图文类型：下载图片（最多3张）

何时使用

在需要下载抖音链接的内容时使用此技能：

• 下载抖音链接的视频 (v.douyin.com/xxx)
• 下载抖音链接的图文
• 解析视频信息（标题、作者、统计数据等）

前置要求

• Node.js 环境

使用方法

解析并下载

node ~/.openclaw/skills/douyin-downloader-js/script/douyin-parse.cjs "抖音链接" [输出目录]

• 第一个参数：抖音链接（必填）
• 第二个参数：输出目录（可选，默认 ./output）

自动识别

脚本会自动识别内容类型：

• 视频类型：返回 videoUrl，下载视频
• 图文类型：返回 images 数组，下载图片

图文处理规则

• 图文数量 ≤ 3张：全部发送
• 图文数量 > 3张：只发送前3张，告知用户剩余数量
• 示例：共7张图片 → 发送前3张 → 提示"还有4张图片未发送"

返回格式

视频类型

{
  "parseId": "视频ID",
  "title": "视频标题",
  "author": { "nickname": "作者昵称", "avatar": "头像URL" },
  "statistics": { "digg": 点赞数, "comment": 评论数, "share": 分享数 },
  "images": [],
  "cover": "封面URL",
  "videoUrl": "视频URL（去除水印）"
}

图文类型

{
  "parseId": "图文ID",
  "title": "图文标题",
  "author": { "nickname": "作者昵称", "avatar": "头像URL" },
  "statistics": { "digg": 点赞数, "comment": 评论数, "share": 分享数 },
  "images": ["图片URL1", "图片URL2", ...],
  "cover": "封面URL",
  "videoUrl": ""
}

技术细节

• 使用移动端 User-Agent 模拟手机访问
• 从 window._ROUTER_DATA 中提取数据
• 自动去除视频水印（playwm → play）
• 短链接自动重定向到完整URL

注意事项

• 视频/图片 URL 是临时的，可能会过期
• 某些内容可能需要登录才能访问
• 下载完成后删除本地临时文件
• 下载的文件默认保存到你的工作空间