Turn any public YouTube video into ready to use marketing notes. It gathers summaries, transcripts, timestamps, scenes, or clip ideas, then delivers the result to your email and saves a file to your Drive. This is helpful for marketers and creators who need fast content from video.

A simple form starts the run with a video ID and the output type you want. The flow pulls video details from the YouTube Data API, builds an audience profile prompt, and asks the Google Generative Language API for structured metadata. It then fills one of several prompt templates and calls the same API again to create clean markdown content. The content converts to HTML, sends by Gmail, stores as a text file in Google Drive, and also returns an HTML view to the form. Merge logic keeps video data and AI results in one record.

You will need a Google Cloud API key with access to the Generative Language API and the YouTube Data API, plus OAuth for Gmail and Google Drive. Most teams see work time drop from hours of manual notes to minutes per video, with more consistent output. Use it for social post copy, show notes, course outlines, and clip planning across many videos.

• Form based trigger collects a YouTube video ID and the output type you want.
• YouTube Data API fetches video details to give context for the content.
• Audience metadata prompt guides the AI to focus on the right reader.
• Prompt templates cover default summary, full transcript, timestamps, scene description, and clip ideas.
• Code step selects the correct template by tag based on your chosen prompt type.
• Google Generative Language API creates structured markdown content.
• Markdown converts to HTML for clean email and web display.
• Gmail sends the HTML result to your inbox for quick review.
• Google Drive saves a text file copy for organized storage and search.
• Merge keeps video details and AI output together for easy tracking.

• Reduce manual work from 2 hours to 10 minutes per video
• Automate up to 90% of summarization and transcription tasks
• Connect four Google services in one seamless flow
• Handle 10 times more videos per week without extra staff
• Use 100% standard templates for consistent outputs

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 Drive, Gmail, Google Generative Language API and YouTube Data API. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, open the Gmail node, choose Credential to connect with, click Create new credential, and follow the on screen steps to connect your Gmail account with OAuth2.
4. In the n8n credentials manager, open the Google Drive node, choose Credential to connect with, click Create new credential, and follow the on screen steps to connect your Google Drive account with OAuth2.
5. In Google Cloud Console, enable YouTube Data API v3 and the Generative Language API. Create an API key and copy it.
6. In n8n Cloud, add an environment variable named GOOGLE_API_KEY with your API key. Alternatively, open the Config node and paste your key into the google_api_key field.
7. Open the Create YouTube API URL and Get YouTube Video Details nodes. Confirm the API key is used in the request (query string or header). Save changes.
8. Open the Start Workflow form trigger. Copy the public form URL. Make sure it has fields for Video ID and Prompt Type.
9. Review the Compose Prompts node. Adjust the templates for default, transcribe, timestamps, summary, scene, and clips if you want a different tone.
10. Run a test: open the form URL, enter a public YouTube video ID, pick an output type, and submit. Check the execution in n8n to confirm each node runs without errors.
11. Verify delivery: look for the HTML email in Gmail and the saved text file in Google Drive. If you see 403 or 400 errors from Google APIs, check that the APIs are enabled and the key is correct.
12. If the AI response format changes, open the Extract MetaData Object node and confirm it parses the response text into JSON as expected.