Skill

Review

Audit score 70

douyin-video

yzfly/douyin-mcp-server

Download watermark-free Douyin videos and extract captions via speech recognition.

What is douyin-video?

A tool for processing Douyin (Chinese TikTok) video links. It retrieves watermark-free download URLs, downloads videos locally, and extracts speech-to-text captions automatically saved to files. Use this when you need to work with Douyin video content, batch process videos, or extract transcripts.

Parse Douyin share links and extract watermark-free video download URLs without API keys
Download videos to local directories with progress tracking
Extract speech-to-text captions from videos using SiliconFlow API and save as Markdown files
Organize extracted captions and videos in folder structures by video ID
Support batch processing and quiet mode for reduced output

How to install douyin-video

npx skills add https://github.com/yzfly/douyin-mcp-server --skill douyin-video

Prerequisites

FFmpeg installed on system (macOS: brew install ffmpeg; Ubuntu: apt install ffmpeg)
Python with requests and ffmpeg-python packages (pip install requests ffmpeg-python)
SiliconFlow API key set as API_KEY environment variable (only required for caption extraction)

Claude Code

Cursor

Windsurf

Cline

How to use douyin-video

1.Install FFmpeg on your system using your package manager
2.Install Python dependencies: pip install requests ffmpeg-python
3.Set the API_KEY environment variable with your SiliconFlow API key (if extracting captions)
4.Run the script with a Douyin share link and desired action (info, download, or extract)
5.Check the output directory for downloaded videos and extracted captions in Markdown format

Use cases

Good for

Download watermark-free Douyin videos for archival or analysis
Extract and organize video transcripts from multiple Douyin videos
Automate caption extraction for content creators or researchers
Retrieve video metadata (ID, title, download link) from Douyin share links
Process Douyin content in bulk with command-line or Python API

Who it's for

Content researchers and analysts
Social media content creators
Developers building Douyin automation tools
Users needing to archive or process Chinese short-form video content

douyin-video FAQ

Do I need an API key to download videos?

No. Getting video info and downloading watermark-free videos requires no API key. Only caption extraction via speech recognition requires a SiliconFlow API key.

What Douyin link formats are supported?

The tool supports standard Douyin share links (typically https://v.douyin.com/xxxxx/) and full Douyin video URLs. Ensure the link is valid and accessible.

How are extracted captions organized?

Each video's caption is saved in a folder named by its video ID, with a transcript.md file in Markdown format containing video metadata and the extracted text.

What should I do if caption extraction fails?

Verify your API_KEY environment variable is set correctly and has sufficient quota, ensure FFmpeg is properly installed, and confirm the video link is valid.

Can I use this tool for commercial purposes?

The tool is intended for learning and research use only. Ensure compliance with applicable laws and do not use for copyright infringement or illegal purposes.

Full instructions (SKILL.md)

Source of truth, from yzfly/douyin-mcp-server.

name: douyin-video description: "抖音无水印视频下载和文案提取工具. 从抖音分享链接获取无水印视频下载链接, 下载视频, 提取视频中的语音文案并自动保存到文件. 适用场景包括获取抖音视频信息, 下载无水印视频, 批量提取视频文案. 当用户需要处理抖音视频链接或提取视频内容时触发."

抖音无水印视频下载和文案提取

从抖音分享链接获取无水印视频下载链接, 下载视频, 并使用语音识别提取视频中的文案, 自动保存到文件.

功能概述

获取下载链接: 从抖音分享链接解析出无水印视频的直接下载地址 (无需 API 密钥)
下载视频: 将无水印视频下载到本地指定目录
提取文案: 通过语音识别从视频中提取文字内容 (需要硅基流动 API 密钥)
自动保存: 每个视频的文案自动保存到独立文件夹 (视频ID为文件夹名)

环境要求

依赖安装

pip install requests ffmpeg-python

系统要求

FFmpeg 必须安装在系统中 (用于音视频处理)
macOS: brew install ffmpeg
Ubuntu: apt install ffmpeg

API 密钥配置 (仅文案提取需要)

文案提取功能使用硅基流动 API, 需要设置环境变量:

export API_KEY="your-siliconflow-api-key"

获取 API 密钥: https://cloud.siliconflow.cn/

使用方法

方法一: 使用脚本 (推荐)

# 获取视频信息和下载链接 (无需 API 密钥)
python douyin_downloader.py --link "抖音分享链接" --action info

# 下载视频到指定目录
python douyin_downloader.py --link "抖音分享链接" --action download --output ./videos

# 提取视频文案并保存到文件 (需要 API_KEY 环境变量)
python douyin_downloader.py --link "抖音分享链接" --action extract --output ./output

# 提取文案并同时保存视频
python douyin_downloader.py --link "抖音分享链接" --action extract --output ./output --save-video

# 安静模式 (减少输出)
python douyin_downloader.py --link "抖音分享链接" --action extract --output ./output --quiet

输出目录结构

提取文案后, 每个视频会保存到独立文件夹:

output/
├── 7600361826030865707/      # 视频ID为文件夹名
│   └── transcript.md         # Markdown 格式文案文件
├── 7581044356631612699/
│   ├── transcript.md
│   └── 7581044356631612699.mp4  # 使用 --save-video 时保存
└── ...

Markdown 文案格式

# 视频标题

| 属性 | 值 |
|------|-----|
| 视频ID | `7600361826030865707` |
| 提取时间 | 2026-01-30 14:19:00 |
| 下载链接 | [点击下载](url) |

---

## 文案内容

(语音识别的文字内容)

方法二: 在 Python 代码中调用

from douyin_downloader import get_video_info, download_video, extract_text

# 获取视频信息
info = get_video_info("抖音分享链接")
print(f"视频ID: {info['video_id']}")
print(f"标题: {info['title']}")
print(f"下载链接: {info['url']}")

# 下载视频
video_path = download_video("抖音分享链接", output_dir="./videos")

# 提取文案并保存到文件
result = extract_text("抖音分享链接", output_dir="./output")
print(f"文案已保存到: {result['output_path']}")
print(result['text'])