Turn any public YouTube link into useful content on demand. The flow can create a clean transcript, a timestamped transcript, a scene description, short clip ideas, or a summary for blogs and video SEO. It is ideal for marketing and content teams that want fast, repeatable outputs from videos.

The run starts with a manual trigger and sets core fields like automation ID, API key, model, prompt type, and the video URL. A switch chooses the right prompt based on the prompt type you set, then packages the request. The HTTP Request node calls the Google Generative Language API with your prompt and the video URL. A small code step merges the API response with earlier fields, and a mapping step extracts the final answer so it is easy to pass downstream. An optional error path helps you add custom handling if the API returns an issue.

Setup is simple. You only need a Google API key and a public YouTube URL. Paste the API key once and choose the output you want. Expect to reduce manual transcription and packaging from hours to minutes. Common uses include full YouTube SEO packages, timestamped notes for editing, social clip ideas, and concise blog summaries that match the video.

• Manual trigger to run ad hoc video analysis on demand.
• Set node defines API key, model, YouTube URL, and prompt type in one place.
• Switch node routes to the right prompt for transcript, timestamps, summary, scene, or clip ideas.
• Multiple Set nodes store ready to use prompts so you can swap outputs with one field.
• HTTP Request calls the Google Generative Language API with the prompt and video URL.
• Code node merges API results with prior fields to keep context intact.
• If node detects errors so you can add custom handling.
• Final Set node extracts the answer text for easy use in next steps.

• Reduce manual transcription from 2 hours to 5 minutes per video
• Automate 80% of YouTube SEO packaging work including titles, descriptions, keywords, and timestamps
• Improve consistency of summaries and clips by using the same prompts every time
• Scale video analysis to 10 times more videos without adding staff
• Connect insights to any downstream tool after the final mapping step

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Google Generative Language API. See the Tools Required section above for links to create accounts with these services.
3. In Google AI Studio, create an API key for the Generative Language API and copy it.
4. Open the Set: Define Initial Variables node and paste your API key into the apiKey field. Set model to gemini 1.5 flash and paste your public YouTube URL into youtubeURL.
5. In the same node, choose your promptType. Supported values include Transcript, Transcript with Timestamps, Summarize, Scene, and Scene Clips.
6. Optional: Open each Set node that holds a prompt and adjust wording to match your brand or output format.
7. Double click the HTTP Request to Google node. In the Credential to connect with dropdown, click Create new credential. If asked for authentication, choose None because the API key is sent in the URL. Save the credential.
8. Click Execute workflow to run a test. Use a short public video first to verify the end to end call completes.
9. Open the HTTP Request to Google node output. If you see an error field, check the API key, model name, and that the video URL is public.
10. Check the Set Fields: Define Outcome node. Confirm the answerAIGenerated field contains the final text you expect.
11. Security tip: Store the API key in n8n credentials or environment variables and avoid hardcoding in shared templates.