Converting Youtube videos to X post

Get Help
1

I’ve been having some issues building a flow that can do the following:

  1. Continually check a specific YouTube channel for new videos.
  2. When a new video is posted, the flow will extract the transcript.
  3. The transcript will be summarised.
  4. The transcription will be converted to an X post or thread
  5. The X post or thread will be automatically posted to X

I have been able to build a flow from one of the templates but the biggest problem I am having is fetching new videos from the particular YouTube channel. I have used the Google alerts RSS reader but it does not seem to work. I have created a custom alert with Google as per the instructions but information about newly released videos are not being fed into the flow.

2

Hey @Thomas_Walker! If you’re reporting an issue with a flow or an error in a run, please include the run link and make sure it’s shareable so we can take a look.

  1. Find your run link on the history page. Format: https://www.gumloop.com/pipeline?run_id={{your_run_id}}&workbook_id={{workbook_id}}

  2. Make it shareable by clicking “Share” → ‘Anyone with the link can view’ in the top-left corner of the flow screen.
    GIF guide

  3. Provide details about the issue—more context helps us troubleshoot faster.

You can find your run history here: https://www.gumloop.com/history

3

https://www.gumloop.com/pipeline?run_id=288F3SkFWKURhUMbWw8iBX&workbook_id=cqxbNzUZyPL2VubVc3YzmN

4

https://www.gumloop.com/pipeline?workbook_id=cqxbNzUZyPL2VubVc3YzmN&tab=1&run_id=VVr5cQmrfa9c7YgzvSaBDB

5

Hey @Thomas_Walker — I think parts 3 to 5 are spot on, exactly how I would’ve approached it too.

The main issue is that automations or workflows don’t really have memory — so they don’t actually know what YouTube video is “new.”

A relatively straightforward fix could be to scrape the YouTube channel using either the Web Agent Scraper node or the Website Scraper node. You’d extract the YouTube links using the Extract Data node — that becomes one subflow.

Then, at the end of your main flow, you’d also post the video links you’ve processed into a spreadsheet.

Next, use a Run Code or Custom node to compare the two lists: the “existing YouTube links” from scraper subflow and the “processed YouTube links” from your spreadsheet. Here’s a video tutorial for custom nodes that might help.

By comparing the two, you can spot which links are new (if any) and process only those.

If you schedule this to run daily, it would:

  1. Scrape YouTube
  2. Extract all video links
  3. Feed them into the comparison node alongside the list of processed links
  4. If a link is new, it runs the rest of the flow
  5. Then logs it as processed in the spreadsheet

It might sound a bit involved at first, but once you start building it out, it should come together pretty smoothly.

1 Like