yby6-video-parser

此技能用于解析短视频链接，支持解析抖音、快手、B站等多个主流平台的短视频和图文链接，并能自动提取语音内容转录为文字。适用于需要批量获取视频元数据或将视频内容转为文本的场景时使用此 skill。

Security Scan

The skill's code generally matches its video‑parsing/transcription purpose, but the package metadata omits the sensitive environment variables it actually uses and the runtime instructions allow sending links/audio to external services — the mismatches merit caution before installing or supplying secrets.

Purpose & Capability

Name/description match the included code: many per‑platform parsers, a download/extract/transcribe pipeline, and output to demos/tmp. That capability set reasonably explains the included modules. However, the registry metadata declares no required environment variables or primary credential while the README/SKILL.md and the code clearly expect a SiliconFlow API key for transcription (and optionally an external parse_api_url). This omission in metadata is an incoherence and reduces transparency.

Instruction Scope

SKILL.md and scripts instruct the agent to parse arbitrary share URLs, download remote video files, run ffmpeg locally to extract audio, call an external ASR API (SiliconFlow) with the audio, and write results to demos/ and tmp/. The code also supports using a user‑supplied parse_api_url (an external parsing API). These behaviors are expected for the stated purpose but are broad: they involve network access, file writes, subprocess invocation (ffmpeg), and transmitting audio/URLs to remote services. The skill will read a .env file for secrets even though the registry declared none.

Install Mechanism

There is no automated install step in the skill metadata; it's an instruction+source bundle. The repository contains a requirements.txt (httpx, fake-useragent, requests) and instructs users to pip install them — standard and proportional. No remote download/installation of arbitrary archives or shorteners was found.

Credentials

The skill requires an API key for SiliconFlow (api_key) to perform transcription and reads other optional settings from a .env file (parse_api_url, siliconflow_api_url, auto_cleanup_temp_files). The registry lists no required env vars or primary credential, which is inconsistent. Supplying the API key gives a third‑party service access to audio extracted from any parsed video; parse_api_url (if set) would forward URLs to an arbitrary endpoint. These environment/credential requirements are meaningful and should have been declared in metadata.

Persistence & Privilege

The skill does not request always:true and does not modify other skills or system configs. It creates files under tmp/ and demos/ and runs local ffmpeg subprocesses — normal for this functionality. Model invocation is allowed (default) but not combined here with other elevated privileges.

What to consider before installing

This package generally does what it says (parses many short‑video platforms and can transcribe audio), but there are important transparency issues: the repository expects a SiliconFlow API key (and can call an external parse_api_url) even though the registry metadata lists no required credentials. Before installing or entering secrets: 1) Do not provide your SiliconFlow API key unless you trust that service and the skill maintainer; the key will be used to upload audio to SiliconFlow for transcription. 2) Do not set parse_api_url to an untrusted endpoint — that would send share URLs to that server. 3) Run the code in a sandboxed environment if possible: it downloads remote videos, writes files under tmp/ and demos/, and invokes ffmpeg via subprocess. 4) Review the code (scripts/transcribe.py and parser modules) yourself or ask the author to explain why metadata omitted required env vars; request an explicit declaration of required credentials. 5) If you only need parsing (no transcription), you can use parse_video_by_url_sync and avoid supplying an API key to limit outbound data. If you want me to, I can highlight exact lines that read .env, perform network calls, or call SiliconFlow so you can review them quickly.

视频解析与转录技能

此技能提供两个核心功能：视频元数据解析和视频语音转录。支持 20+ 个主流视频平台。

使用场景

当出现以下场景时，使用此 skill：

• 当用户需要解析图集内容
• 当用户提供视频分享链接，需要提取视频信息（无水印链接、封面、标题、作者等）时
• 需要将视频中的语音内容转换为文字，用于字幕生成、内容笔记时
• 解析图集内容时

核心功能

功能一：视频元数据解析

使用内置的解析器直接从视频分享链接中提取视频信息，无需依赖外部 API。

功能二：视频语音转录

通过自动化流程：解析视频 → 下载视频 → 提取音频 → 语音转录 → 生成 Markdown 报告。

如何使用

1. 安装依赖

pip install -r requirements.txt

2. 环境配置（仅语音转录功能需要）

复制一份项目根目录当中的.env.example 为 .env 文件：

# SiliconFlow API Key (必填，用于视频转录功能)
# 获取地址: https://siliconflow.cn/
api_key=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

# 视频转录模型 (可选，默认: FunAudioLLM/SenseVoiceSmall)
# 支持的模型列表: https://docs.siliconflow.cn/api-reference/audio
model=FunAudioLLM/SenseVoiceSmall

# 视频解析 API 地址 (可选，用于转录功能)
# 如果留空或注释掉，则使用项目内置的本地解析器（推荐）
# 如果填写了外部 API 地址，则使用外部 API 进行视频解析
# 示例: parse_api_url=http://ip:8000/video/share/url/parse?url=
parse_api_url=

# SiliconFlow ASR API 地址 (可选，默认: https://api.siliconflow.cn/v1/audio/transcriptions)
siliconflow_api_url=https://api.siliconflow.cn/v1/audio/transcriptions

# 是否自动清理临时文件 (可选，默认: false)
# true: 自动删除临时文件（视频和音频）
# false: 保留临时文件在 tmp/目录中
auto_cleanup_temp_files=false

脚本使用说明

脚本一：scripts/skill.py - 视频元数据解析

作用: 解析视频分享链接，获取视频详细信息。

运行方式:

python scripts/skill.py [选项]

参数说明:

参数	说明	必需
--url	要解析的视频分享链接	否
--list_platforms	列出所有支持的平台	否

常用示例:

# 1) 列出支持的平台
python scripts/skill.py --list_platforms

# 2) 通过分享链接解析视频
python scripts/skill.py --url "https://v.douyin.com/xxxxxx"

输出格式:

{
  "video_url": "https://example.com/video.mp4",
  "cover_url": "https://example.com/cover.jpg",
  "title": "视频标题",
  "music_url": "https://example.com/music.mp3",
  "images": [
    {
      "url": "图片链接",
      "live_photo_url": "LivePhoto链接"
    }
  ],
  "author": {
    "uid": "作者ID",
    "name": "作者昵称",
    "avatar": "https://example.com/avatar.jpg"
  }
}

脚本二：scripts/transcribe.py - 视频语音转录

作用: 解析视频链接并自动转录语音内容为文字。

运行方式:

python scripts/transcribe.py --url <VIDEO_URL> [选项]

参数说明:

参数	说明	默认值
--url	必需。要解析的视频分享链接	无
--api_key	SiliconFlow API 密钥。未提供则从 .env 读取	从 .env 读取
--model	语音识别模型名称	FunAudioLLM/SenseVoiceSmall
--parse_result	已解析的结果 JSON 字符串（可选，跳过解析步骤）	无
--auto_cleanup	是否自动清理临时文件（true/false）	从 .env 读取
--use_local_parser	是否使用本地解析器（true/false）	从 .env 读取

常用示例:

# 1) 基础解析与转录
python scripts/transcribe.py --url "https://v.douyin.com/xxxxxx"

# 2) 指定 API Key 和模型
python scripts/transcribe.py --url "https://www.xiaohongshu.com/explore/xxxx" --api-key sk-your-key --model FunAudioLLM/SenseVoiceSmall

# 3) 保留临时文件
python scripts/transcribe.py --url "https://www.bilibili.com/video/xxxx" --auto_cleanup false

输出:

• 命令行实时打印 JSON 格式结果
• 自动在 demos/ 目录下生成结构化的 Markdown 报告

技能资源

核心脚本

• scripts/skill.py - 视频元数据解析脚本
• scripts/transcribe.py - 视频语音转录脚本

解析器目录

• scripts/parser/ - 包含 20+ 个平台的解析器实现

输出目录

• demos/ - Markdown 报告输出目录
• tmp/ - 临时文件存储目录（视频、音频）

支持的平台

平台	标识符	平台	标识符
抖音	douyin	快手	kuaishou
小红书	redbook	哔哩哔哩	bilibili
微博	weibo	皮皮虾	pipixia
西瓜视频	xigua	微视	weishi
绿洲	lvzhou	最右	zuiyou
度小视	quanmin	梨视频	lishipin
皮皮搞笑	pipigaoxiao	虎牙	huya
A站	acfun	逗拍	doupai
美拍	meipai	全民K歌	quanminkge
六间房	sixroom	新片场	xinpianchang
好看视频	haokan	Twitter	twitter

注意事项

1. 环境依赖: 语音转录功能需要预先安装 ffmpeg 并添加到环境变量
2. 网络要求: 脚本需要访问网络进行视频解析和 API 调用
3. API 限制: 语音转录受 SiliconFlow 额度限制
4. 输出路径: 临时文件默认保存在 tmp/ 目录，Markdown 报告保存在 demos/ 目录
5. 解析模式: 默认使用项目内置的本地解析器，无需配置外部 API

本 Skill 基于 parse-video-py 项目重构而来，感谢原作者的贡献。