Built with n8n: a YouTube channel that makes and publishes its own videos, plus a live stream steered by chat (fail-closed moderation)

I run a small YouTube character channel, Cynic Stick, on n8n. This is a walkthrough of how it is wired, with the parts I think are reusable. Much of it was built with Claude through the n8n MCP server: I direct, it builds.

The daily and weekly pipeline (about 25 active workflows)

  • Research: scheduled trend and signal searches feed a research workflow. An Ollama research workflow is the fallback.
  • Production: one workflow writes and renders the daily short. The weekly long video is split into script and scene prompts, scene render, beat render, compile and review.
  • Model registry: one small workflow returns a fixed set of models (writer, medium, lite, image, speech-to-text, text-to-speech). Every other workflow asks it, so a model change is one edit.
  • Publishing: separate workflows for YouTube, X, Instagram and Facebook on a schedule. TikTok uploads to drafts and sends me the caption on Telegram for a one-tap manual post.
  • Watchdog: 30 minutes before the publish time, a workflow checks that the build started and finished. It messages me only when it did not, so silence means healthy.
  • Housekeeping: a daily workflow sorts published shorts into 5 YouTube playlists with a light model. A site export and a chat agent keep cynicstick.com and its “Ask Cynic Stick” widget in sync with the channel.

The new part: a live stream where chat steers the story (12 active workflows)
Viewers type !scene plus an idea (or !ask plus a question) in the YouTube live chat. The stream is an ASCII-art world drawn by a web page that reads an n8n endpoint.

  • Poller: reads the live chat on a configurable interval and stays idle unless the automation is on and not locked.
  • Ingest: de-duplicates, allows one idea per slot, then calls the moderation sub-workflow.
  • Moderation: name and rules checks, then two text judges and three art judges through OpenRouter with strict JSON output. Anything unclear is rejected (fail closed). A model turns the idea into a validated scene, and its output is checked again.
  • Human veto: every accepted idea goes to a Telegram message with a veto button. When the window ends, a release workflow closes the message and announces the idea in the chat. Repeated vetoes lock the automation.
  • Safety nets: a kill switch (panic and resume webhook), an hourly canary that sends four fixed lines through the judges and locks the automation if a clearly bad one passes, a keeper that checks every minute that the broadcast is live, and a daily retention job that removes raw comment text after 7 days and names and channel IDs after 30.

What I would like feedback on

  1. Is an hourly canary enough to trust fail-closed judges, or do you test moderation differently?
  2. How do you balance chat polling latency against YouTube API quota?
  3. Anything in the retention approach you would change?

Channel: https://www.youtube.com/@cynicstick
Site: https://cynicstick.com
The live is brand new and the audience is tiny, so I care about the design more than the numbers.

1 Like

Hi Sam,

I went through your system and noticed a couple of gaps. Not bugs, just things I think are missing.

Right now, when something fails, the watchdog tells you about it. That’s good, but I didn’t see a way for the pipeline to recover by itself: it reports the error, and then someone has to step in.

Since you already split the weekly video into stages, I would save a checkpoint at the end of every stage (script done, scenes rendered, video compiled, and so on). When something fails, the run continues from the last checkpoint instead of starting over.

I’d also sort errors by type:

  • Quota errors: find out when each service resets its quota and wait until then. For example, the YouTube Data API quota resets at midnight Pacific Time. When a quota error comes in, the workflow waits until the reset and then continues on its own.
  • Temporary errors (timeouts, 5xx): retry after a few minutes.
  • Anything else: stop, send the details to Telegram, and resume from the checkpoint once it’s fixed.

This also connects to your second question about API quota. API calls cost money and time. With checkpoints, a failure means redoing one step at most, not the whole flow. Which makes more sense: rerunning the whole pipeline, or one stage?

It’s your call, of course, I’m just sharing an idea. It’s a really impressive project. I’ve been trying to build something similar for months without much success, so this was inspiring to read. Thanks for sharing it.

Ömer

Thanks Ömer, useful ideas, and thank you for the kind words.

I went through my workflows against your list. Transient errors are already retried (the render nodes try 5 times, 5 seconds apart), and the live chat poller sorts errors by type: 5xx and timeouts back off from 30 seconds to 10 minutes, quota and auth errors from 2 to 30 minutes, with its state kept in a data table. The render stage also has a checkpoint: scenes already finished on disk are skipped when a job is re-run.

Where you are right: after a failure the re-run is manual, nothing waits for a quota or credit reset (a 402 just stops and alerts me on Telegram), and I could not find an automatic resume across the whole weekly chain of script, render and compile. That is the part I will work on.

1 Like