ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified media Safety 4/5

video-watcher

Fetch and read transcripts from YouTube and Bilibili videos. Use when you need to summarize a video, answer questions about its content, or extract information from it.

Why use this skill?

Use OpenClaw's video-watcher skill to fetch and summarize YouTube and Bilibili transcripts. Effortlessly turn video content into searchable, actionable text.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/donnycui/bilibili-youtube-watcher
Or

What This Skill Does

The video-watcher skill is a powerful utility within the OpenClaw ecosystem designed to bridge the gap between long-form video content and text-based analysis. By integrating with the robust yt-dlp library, this skill allows users to programmatically fetch, parse, and retrieve transcripts from both YouTube and Bilibili. Rather than forcing users to watch hours of content to find a specific answer or summarize key points, this skill extracts the underlying subtitle or closed-caption data, turning visual media into searchable, processable text for the AI agent.

Installation

To integrate the video-watcher into your environment, ensure you have yt-dlp installed and accessible in your system's PATH. Once your environment is prepared, execute the following installation command in your terminal:

clawhub install openclaw/skills/skills/donnycui/bilibili-youtube-watcher

This command automatically handles the dependency resolution for the skill, allowing you to invoke the get_transcript.py script directly from your OpenClaw agent workflows.

Use Cases

This skill is indispensable for professionals, researchers, and content creators who need to digest video data rapidly. Common use cases include:

  • Educational Research: Extracting lecture notes or transcriptions from academic videos on YouTube.
  • Market Analysis: Analyzing Bilibili trends or product reviews to extract sentiment and feature lists without manual playback.
  • Content Repurposing: Turning long-form video interviews into blog posts, newsletters, or social media summaries by processing the transcript first.
  • Quick Fact-Checking: Using the AI agent to search for specific claims or timestamps within a video, saving significant time during technical investigation.

Example Prompts

  1. "Can you watch this YouTube video [URL] and provide a 5-bullet point summary of the key takeaways for me?"
  2. "Please check the Bilibili video [URL] and tell me what the speaker says about the new software release, specifically looking for any mentions of performance benchmarks."
  3. "Fetch the transcript for [URL] in English. If it is not available, try searching for the Simplified Chinese version and summarize the main arguments."

Tips & Limitations

  • Subtitles Required: This skill relies on the presence of existing closed captions or auto-generated subtitles. If a video creator has disabled subtitles, the script will be unable to generate a transcript.
  • Language Precision: Always specify the --lang flag if you require a specific version of a transcript. By default, it falls back to English for YouTube and Chinese for Bilibili.
  • Network Performance: Ensure your internet connection is stable, as fetching subtitles for very long videos may take a moment to process.
  • Dependency Maintenance: Keep your yt-dlp version updated to ensure continued compatibility with YouTube and Bilibili's frequently changing web structures.

Metadata

Author@donnycui
Stars2387
Views1
Updated2026-03-09
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-donnycui-bilibili-youtube-watcher": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#video-transcription#youtube#bilibili#summarization#nlp
Safety Score: 4/5

Flags: network-access, code-execution