LinkedIn video extraction

The user-facing part.

Extract transcripts and metadata from LinkedIn video posts. Uoink uses its local yt-dlp path to fetch video streams, generates transcripts, and writes clean markdown. It files the output by topic, saving hours of manual notes.

Run the loop.

1. Right-click any video post on LinkedIn.
2. Click Uoink video in the browser context menu.
3. Monitor progress in the extension toolbar popup.
4. Check your library folder for the auto-filed transcript.
5. Open the markdown file to read it as a document.

How Uoink handles it.

The helper uses its local video tools to query the LinkedIn post. It extracts the raw media streams, routes them through the local Whisper transcriber, and builds the markdown metadata. The resulting files are saved to your local directory and indexed in the search database.

Agent-readable surface.

{
  "name": "uoink_video",
  "arguments": {
    "url": "https://www.linkedin.com/posts/activity-123456789"
  }
}

Full server metadata lives at /mcp and /mcp/manifest.json.

Run it on your next source.

Convert LinkedIn video posts into readable local text documents.

Install Uoink