Generate AI B-roll clips from videos with Veo 3

by Koulikas Giannis ✓ · 24 views · created 2026-04-19 · Content Creation, Multimodal AI

What this workflow does

This AI workflow has 25 nodes and runs manually, when you click “Execute workflow”. Along the way it connects JWT, OpenAI, Google Cloud Storage and Telegram.

How it runs

Trigger: Manually, when you click “Execute workflow”.

1. Set Video URL (Edit Fields (Set)): sets or maps fields
2. Set API Credentials (Edit Fields (Set)): sets or maps fields
3. Generate JWT Token (JWT): uses JWT
4. Request OAuth Token (HTTP Request): calls an HTTP API
5. Fetch Video from URL (HTTP Request): calls an HTTP API
6. Extract Data from File (Extract from File): extracts data from a file
7. Prepare Analysis Prompt (Code): runs custom JavaScript code
8. Post Video for Analysis (HTTP Request): calls an HTTP API
9. Select Key Moments with AI (OpenAI): uses OpenAI
10. Split Moments into Items (Code): runs custom JavaScript code

…plus 14 more steps. See the full node list below.

Services it uses

JWTOpenAIGoogle Cloud StorageTelegram

Before you upgrade to n8n 3.0

Nodes used (14)

• Edit Fields (Set)n8n-nodes-base.set
• JWTn8n-nodes-base.jwt
• HTTP Requestn8n-nodes-base.httpRequest
• Extract from Filen8n-nodes-base.extractFromFile
• OpenAI@n8n/n8n-nodes-langchain.openAi
• Coden8n-nodes-base.code
• Loop Over Items (Split in Batches)n8n-nodes-base.splitInBatches
• Switchn8n-nodes-base.switch
• Waitn8n-nodes-base.wait
• Convert to Filen8n-nodes-base.convertToFile
• Google Cloud Storagen8n-nodes-base.googleCloudStorage
• Aggregaten8n-nodes-base.aggregate
• Telegramn8n-nodes-base.telegram
• Manual Triggern8n-nodes-base.manualTrigger

From the author

How it works 1. Analyze the video – Downloads a 2-minute video and uses Gemini 2.5 Flash to identify emotional peaks, hook moments, and timestamps 2. Select best moments – Uses GPT-4o-mini to pick the 3 best moments for B-roll insertion based on confidence…

View the original on n8n.io → Check your own copy