Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Analyzes video content from URLs by downloading, transcribing audio locally, and answering questions about the content. Works with TikTok, YouTube, Instagram, Twitter/X, and 1000+ other platforms. Transcripts are cached so follow-up questions don't require re-downloading.
Transcription runs locally using faster-whisper, so no audio is sent to external APIs and processing works offline after the initial model download.
When to use it
- Summarize a TikTok tutorial without watching it
- Extract key points from a YouTube lecture
- Check what a viral video is actually saying
- Ask follow-up questions about a previously saved video
- Analyze an Instagram reel for its hook and structure
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'What is this TikTok teaching?' with a tiktok.com URL
- 1Send immediate acknowledgment message to user
- 2Check if transcript is cached; if not, download video with yt-dlp
- 3Send progress update, then transcribe audio locally using faster-whisper
- 4Answer the user's question using the transcript text
- 5Offer to save the transcript to the local library for future queries
Summary of what the video teaches, key points as a bullet list, and a save prompt
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.