skills.rest
dontbesilent2025
dontbesilent2025COMMUNITY
Explore all skills by this author →
WebsiteGitHub

dbs-video-extract

Extract short-video metadata and speech transcripts from Douyin, Xiaohongshu, and WeChat Channels links.

9.8k|1.1k|Updated Mar 20, 2026
Search Tags:#metadata extraction#speech to text#video transcript#Douyin#tikhub#XiaoHongShu#markdown archive
One-click install
npx skills add https://github.com/dontbesilent2025/dbskill --skill dbs-video-extract
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dbs-video-extract
Source: https://github.com/dontbesilent2025/dbskill/tree/main/skills/dbs-video-extract
Command: npx skills add https://github.com/dontbesilent2025/dbskill --skill dbs-video-extract

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Turning a shared short-video link into usable data normally requires manual copying, third-party tools, and scattered notes. This Skill accepts a Douyin, Xiaohongshu, or WeChat Channels share link or share text, queries structured video/account data via TikHub, extracts the speech transcript via Qingdou, and archives the result as a Markdown file organized by author and title.

Core Features & Use Cases

  • Structured data query: Fetches video or account statistics (likes, comments, shares, follower counts) through TikHub MCP for Douyin, Xiaohongshu, and WeChat Channels, with an App-to-Web fallback for Douyin videos.
  • Speech-to-text transcripts: Submits share text to the Qingdou API, polls for completion, and saves faithful transcripts as Markdown files under {output}/{author}/{title}.md with YAML frontmatter.
  • Credential preflight and onboarding: Runs a read-only --check-keys check, explains API keys in plain language, and guides users through purchasing and securely configuring TikHub or Qingdou credentials without exposing keys in chat.
  • Use Case: A content creator researching competitors pastes a Douyin share text; the Skill returns engagement statistics plus a saved Markdown transcript for later content analysis.

Quick Start

Ask the agent to extract the data and transcript from a short-video share link, for example: "Use dbs-video-extract to get the data and transcript for this video: <paste share text>".

Frequently Asked Questions about dbs-video-extract

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract a transcript from a Douyin or Xiaohongshu video link?▼

Paste the full share text or link and run the extraction script in the default both mode. The Qingdou API converts the speech to text, and the result is saved as a Markdown file under an author/title folder structure.

What video platforms are supported for data queries?▼

TikHub data queries support Douyin (videos and user profiles), Xiaohongshu video notes, and WeChat Channels videos. Transcript extraction via Qingdou may work on additional platforms that Qingdou can parse.

Do I need both TikHub and Qingdou API keys?▼

No. A single key enables partial functionality: TikHub alone provides video and account data, while Qingdou alone provides transcripts. The preflight check reports the credential state and narrows the mode to whichever service is configured.

Where are my API keys stored and are they safe?▼

Keys are read from environment variables, a local config file at ~/.config/dbs/API_Keys.md with owner-only permissions, or the macOS Keychain. The configuration script reads keys via hidden terminal input and never prints them in commands, logs, or chat.

What happens if one service fails during extraction?▼

The script continues with the other service and marks the result as partial_success. Already obtained data or transcripts are preserved, and the JSON summary reports which side failed and why.

Why does the skill skip writing a transcript file that already exists?▼

To protect user edits, existing Markdown files from the same source URL are skipped by default. Pass the --overwrite flag only when you explicitly want to regenerate the transcript.

Related Skills

Explore More Content & Communication Skills→
AradotsoAradotso
7

douyin-video-extractor

Extract watermark-free Douyin/TikTok videos and transcribe audio to Markdown.

Official
Intermediate
vivy-yivivy-yi
363

huitun-data

Analyze Xiaohongshu account performance, track content metrics, and research competitors.

Community
Advanced
pingdiorpingdior
7

prohibited-words-detection

Detect prohibited claims and compliance risks in marketing text.

Community
Intermediate
genesis-plangenesis-plan

video-publish-all

Publish a local video to Xiaohongshu, Douyin, Bilibili, and WeChat Channels with unified metadata.

Community
Intermediate