Douyin Video Processor

Douyin Video Processor

Process Douyin (TikTok China) share links: extract video speech to text, get watermark-free download URLs, and download videos locally.

First-Time Setup

Before using any script, install the Python dependencies:

bash scripts/install_deps.sh

This installs requests and dashscope. Only needed once.

Environment Variable

The text extraction feature requires DASHSCOPE_API_KEY from Alibaba Cloud Bailian.

The user must set this before you can call douyin_extract_text.py. If it is not set, tell the user:

To use text extraction, you need an Alibaba Cloud Bailian API key.

1. Go to https://help.aliyun.com/zh/model-studio/get-api-key
2. Create an API key (starts with sk-)
3. Set it: /secrets set DASHSCOPE_API_KEY sk-your-key

The other two scripts (parse and download) do not need any API key.

Available Scripts

1. Parse Video Info (no API key needed)

Extract video metadata (ID, title, watermark-free URL) from a Douyin share link.

python3 scripts/douyin_parse.py "SHARE_LINK"

Input: A Douyin share link or text containing one (e.g. https://v.douyin.com/xxxxx/ or a full share text like "7.29 复制打开抖音... https://v.douyin.com/xxxxx/")

Output: JSON to stdout:

{
  "status": "success",
  "video_id": "7345678901234567890",
  "title": "Video title here",
  "download_url": "https://..."
}

When to use: User wants to see video info, or you need the download URL without downloading the file.

2. Download Video (no API key needed)

Download a watermark-free video file to local disk.

python3 scripts/douyin_download.py "SHARE_LINK" [output_directory]

Input:

• Arg 1: Douyin share link or text containing one
• Arg 2 (optional): Output directory, defaults to current directory

Output: JSON to stdout:

{
  "status": "success",
  "video_id": "7345678901234567890",
  "title": "Video title here",
  "file_path": "/absolute/path/to/video.mp4",
  "size_bytes": 12345678
}

When to use: User asks to download a Douyin video, save a video, or get the actual video file.

3. Extract Text from Video (requires DASHSCOPE_API_KEY)

Parse a Douyin share link, then transcribe the video speech to text using Alibaba Cloud ASR.

DASHSCOPE_API_KEY="$DASHSCOPE_API_KEY" python3 scripts/douyin_extract_text.py "SHARE_LINK" [model]

Input:

• Arg 1: Douyin share link or text containing one
• Arg 2 (optional): ASR model name, defaults to paraformer-v2

Output: JSON to stdout:

{
  "status": "success",
  "video_id": "7345678901234567890",
  "title": "Video title here",
  "text": "The full transcribed text content from the video..."
}

When to use: User wants to know what's said in a Douyin video, asks to transcribe, extract text, get subtitles, or summarize video content.

Important: Always pass DASHSCOPE_API_KEY as an environment variable in the command. If the key is not set, the script will return an error with setup instructions.

Error Handling

All scripts return JSON even on failure:

{
  "status": "error",
  "error": "Human-readable error message"
}

Common errors and what to tell the user:

Batch Processing

When the user provides multiple Douyin links, process them sequentially. For each link:

1. Call the appropriate script
2. Parse the JSON output
3. Collect results
4. Present a summary table to the user

For batch text extraction, first parse all links with douyin_parse.py to validate them, then extract text only from the valid ones with douyin_extract_text.py. This avoids wasting API calls on broken links.

Cost Awareness

Always prefer douyin_parse.py first when you just need to verify a link or get the download URL.

External Endpoints

Security & Privacy

• DASHSCOPE_API_KEY is only sent to Alibaba Cloud's official API endpoint (dashscope.aliyuncs.com)
• No data is collected, stored, or transmitted by this skill beyond the API calls listed above
• Downloaded video files are saved only where the user specifies
• No persistent background processes

Trust Statement

By using this skill, Douyin share links are sent to Douyin/ByteDance servers for URL resolution, and video URLs may be sent to Alibaba Cloud (Aliyun) for speech-to-text transcription. Only install this skill if you trust these services.