Video Transcript Skill for Openclaw

Extracts subtitles and transcripts from video URLs across multiple platforms and delivers them as formatted .docx files.

vistalyq
v1.0.1
Mar 11, 2026
0
942
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install video2txt

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install video2txt using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Video Transcript Skill?

The Video Transcript Skill is a robust utility designed for AI agents to bridge the gap between video content and text-based documentation. By leveraging the Openclaw Skills framework, this tool automates the extraction of subtitles from major platforms like YouTube and Bilibili, stripping away technical metadata to provide clean, readable text.

This skill goes beyond simple extraction by offering intelligent translation features. For non-Chinese videos, it generates both the original language transcript and a high-quality Chinese translation, making it an essential tool for researchers, students, and content creators who need to digest foreign language media quickly.

Video Transcript Skill Use Cases

  • Creating lecture notes from educational YouTube videos for study and review.
  • Translating Bilibili technical tutorials or variety shows into readable Chinese scripts.
  • Generating text-based archives of webinars and live streams for searchability.
  • Extracting spoken content to assist in blog writing or social media content creation.

How Video Transcript Skill Works

  1. The AI agent detects a video URL or a request for a transcript and initiates the Openclaw Skills workflow.
  2. The system checks for the presence of yt-dlp and the required Node.js docx library.
  3. It uses yt-dlp to fetch the best available subtitle files, preferring manual subtitles over auto-generated ones.
  4. A parsing logic strips SRT timestamps, index numbers, and HTML tags to produce a clean text flow.
  5. If the content is not in Chinese, the agent performs a natural translation into Chinese.
  6. A specialized Node.js script generates professional Word documents (.docx) from the processed text.
  7. The final document paths are provided to the user for immediate access.

Video Transcript Skill Setup

To get started with this skill within the Openclaw Skills ecosystem, ensure your environment is prepared:

# Install yt-dlp for subtitle extraction
pip install yt-dlp

# Navigate to the skill directory and install Node.js dependencies
cd ~/.agents/skills/video-transcript/scripts
npm install

Video Transcript Skill Data Schema & Taxonomy

The skill manages data through a structured lifecycle to ensure clean outputs:

Data Type Description
Temporary Workspace Creates a unique directory in the system TEMP folder for processing.
Source Files Raw .srt files downloaded via yt-dlp are used as the data source.
Refined Text Cleaned plain-text strings merged into paragraphs for readability.
Final Output .docx files containing the original and/or translated content with formatted titles.

Video Transcript Skill Advanced Features

  • Smart subtitle detection: Automatically falls back to auto-generated subtitles if manual ones are unavailable.
  • Multi-platform compatibility: Supports any platform compatible with yt-dlp, including international and domestic sites.
  • Intelligent paragraph merging: Converts fragmented subtitle lines into a cohesive narrative structure.
  • Dual-language delivery: Generates synchronized original and translated documents for comprehensive reference.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*