Emailed report on one YouTube video and its comments

What it is

This n8n workflow turns one YouTube video and its comments into a written report that lands in your inbox. You paste the video’s link in n8n’s chat. A Code node pulls the video ID out of it, and the work splits in two. One branch asks the YouTube Data API for the video’s details, from its title and tags to its statistics and topics, while the other reads every page of top-level comments. Both meet in one item, which an AI agent on gpt-4o-mini reads and writes up in Markdown. The last two nodes turn it into HTML and send it with Gmail.

1. You paste a YouTube link in n8n’s chat
2. The video ID is read from the link
3. YouTube returns the video’s details and its comments
4. An agent on gpt-4o-mini writes an eight-part report
5. Gmail sends the report to your address

Who it suits

Creators who want to know what viewers said under a video, and strategists who report on channels for clients. It reads one video per run. Skip it if you want a view of a whole channel, or if the video has its comments turned off.

Where to use it

Any n8n with a Gmail account it can send from. The video can be any public one, yours or a rival’s, because an API key reads public data only.

• n8n Cloud or your own server: the same file and steps
• n8n’s chat: where you paste each link
• Your inbox: the report comes as an HTML email with the subject YouTube Video Report
• Someone else’s inbox: put a colleague’s or a client’s address in Gmail Report

Setup

Get it saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Its author puts setup at about 13 minutes.

1. In n8n, create a workflow, open its menu, choose Import from File and pick workflow.json.
2. Open Workflow Variables and replace YOUR_YOUTUBE_API_KEY in GOOGLE_API_KEY with your key.
3. Add an OpenAI credential and select it in gpt-4o-mini.
4. Add a Gmail OAuth2 credential and select it in Gmail Report.
5. In the same node, replace YOUR_EMAIL_ADDRESS with the address that should get the report.
6. Save, open the chat, paste a YouTube link and wait for the email.

How to use it

Paste a full YouTube link and nothing else, a youtube.com/watch?v= address or a short youtu.be one. The report comes by email, not in the chat. A video with thousands of comments takes longer, since every page is read before the agent starts.

Example

Say you paste the link of a tutorial with 300 comments. The comment branch reads them in pages and joins them into one field. The agent gets that field with the video’s details, and the email follows the outline its system message sets:

1. Video Overview
2. Comment Analysis
3. Content Opportunities
4. Audience Profile
5. Actionable Recommendations
6. Keywords and Tags
7. Potential Collaborations
8. Detailed Suggestions for Similar Content

For newcomers and experts

If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. This one has 19, and four of them are sticky notes. An API key is the pass Google checks when the workflow asks for public videos and comments. It can read but cannot post or change anything. Send a short video with a few comments first, then open each node to see what it got.

For experts. The comment branch is a Code node that calls commentThreads with this.helpers.request and follows nextPageToken to the last page. It keeps each top-level comment’s textOriginal and nothing else. Replies, likes and dates never reach the agent, though its system message mentions them.

The agent has no tools and no memory.

Good to know

The YouTube key sits in the workflow as plain text, and a big comment section makes a long, costly prompt.

• The key: GOOGLE_API_KEY is a field in Workflow Variables, so anyone who can open or export the workflow can read it
• Long threads: every comment goes to the model in one prompt, so a video with many thousands of comments can pass the model’s limit and fail
• Quota: each page of comments is one YouTube call
• Comments off: on a video with comments turned off, the comment call fails and the run stops there
• Privacy: viewers’ comments go to OpenAI
• The original copy: its Get Video ID code had lost its backslashes, so its fallback pattern could not compile and the node failed on every run. This one has them back
• Credentials: the JSON holds none, and nothing from its author’s accounts