What this workflow does
This workflow gets the newest videos from a YouTube channel.
It reads each video’s transcript and uses AI to create detailed descriptions and tags that help videos show up in searches.
It then updates the video’s metadata on YouTube automatically.
Unpublished videos get set to private and scheduled to publish on Friday at 17:00 UTC.
The main problem solved is saving creators time and reducing mistakes in video metadata editing.
Who should use this workflow
Content creators with many YouTube channels or videos can benefit.
Anyone wanting to cut down on manual title, description, and tag writing should use it.
It fits users who want better search visibility and consistent metadata updates.
Tools and services used
- n8n: Workflow automation platform.
- YouTube API (OAuth2): To fetch video data and update metadata.
- Apify YouTube Transcript Scraper API: To get video transcripts.
- OpenAI GPT-4.1-nano model: To generate video descriptions and titles.
- Google Gemini API (optional): Used here for extra AI tag generation.
Inputs, processing steps, and output
Inputs
- YouTube channel videos fetched via OAuth2 API node.
- Video URLs passed to Apify transcript API with API token.
Processing steps
- Splitting videos to process one by one.
- Getting full video details with YouTube node.
- Setting publish dates and privacy for unpublished videos using JavaScript in a code node.
- Calling Apify API to get full transcript JSON data.
- Running custom JavaScript code to join transcript text into one string.
- Sending transcript text to OpenAI GPT-4 to create long, engaging descriptions in first-person German with hashtags and emojis.
- Sending transcript text to LangChain agent node for German YouTube SEO tag generation.
- Updating video’s metadata (description, tags, optional title) back on YouTube.
Output
- Videos on YouTube with improved, AI-generated descriptions and tags.
- Automatically scheduled publishing for private videos on Fridays 17:00 UTC.
- Time saved from manual metadata edits.
Beginner step-by-step: How to use this workflow in n8n production
Download and import the workflow
- Use the Download button on this page to get the workflow file.
- Open the n8n editor.
- Click the menu, choose “Import from File”, then select the downloaded workflow JSON file.
Configure credentials
- Add your YouTube OAuth2 credentials in the YouTube nodes.
- Enter your OpenAI API Key in the OpenAI nodes.
- Add your Apify API token in the HTTP Request node “Get Transcript” query parameters.
- Provide Google Gemini API credentials in the relevant node if you want extra tag generation.
Check video IDs and URLs
- Confirm the YouTube channel linked is correct in the YouTube nodes.
- If needed, update video or channel IDs in code or node parameters.
Test and activate
- Run the workflow manually using the manual trigger node by clicking “Test workflow”.
- Check outputs for transcript, description, and tag generation.
- Once working, activate the workflow in n8n to run regularly or on schedule.
For advanced hosting options, consider self-host n8n.
Common problems and failures
Sometimes YouTube API calls fail due to missing or expired OAuth tokens.
Check and refresh YouTube OAuth2 credentials inside n8n.
If transcripts come back empty or broken, the video may be private or the token for Apify may be invalid.
Make sure videos are public or unlisted and tokens have correct access.
OpenAI nodes can fail if the prompt has errors or API key limits are reached.
Fix prompt placeholders and verify API quota for OpenAI keys.
Ideas for customization
- Turn on automatic AI title creation by enabling the OpenAI “YT Title” node.
- Replace the Apify transcript API URL and token with other transcript services if preferred.
- Change the JavaScript scheduling code to publish videos on different days or times.
- Modify the “YT Tags” node’s AI prompt to add multilingual tags or niche keywords.
Summary of benefits
✓ Saves time by automating metadata writing from video transcripts.
✓ Improves YouTube SEO with AI-generated descriptions and tags in German.
✓ Reduces errors and keeps video metadata consistent.
✓ Schedules private videos to publish automatically on Fridays at 17:00 UTC.
→ You get faster uploads with better chances to reach the right audience.
Example OpenAI prompt for description generation
The workflow uses a GPT-4.1-nano prompt like this:
Du bist ein professioneller Texteschreiber.
Schreibe eine ausführlichere Zusammenfassung des Videos aus dem folgenden Transkript.
Benutze die Ich-Form, keinen Passiv.
Füge relevante Hashtags und Emojis ein.
Der Text soll auf Deutsch sein.This guides the AI to write in first-person German with SEO tags.