YouTube Comments Agent

Reads every comment your channel collected overnight and posts one morning list: the repeated questions with a draft answer, the business inquiries, and the spam wave you can clear in a single pass.

0 in use1
Loading preview…
Guided setup — test it before connecting anything.

Watch it work before it's live

Run a staged conversation — no account needed. The agent handles it for real while a simulated world answers its tool calls; nothing touches real accounts, and nothing is actually sent.

Comments digest, Tuesday 14 April

Taller de Marta@tallerdemarta

312,000 subscribers. Overnight there are 214 new comments across six videos, 90 of them on Sunday's dovetail jig tutorial, which has passed 180,000 views. Eleven people ask which router bit was used and the video never says, four ask whether the jig fits an imperial base, and one account named for a Spanish tool brand asks who handles paid collaborations.

Set up in minutes

Using this template drops you into a guided setup. It asks exactly this, nothing else:

  1. Connect YouTube

    One sign-in. The agent acts through your account, scoped to what this template uses.

  2. Connect Slack

    One sign-in. The agent acts through your account, scoped to what this template uses.

  3. Channel rules

    What your channel wants surfaced and how you sound when you answer: which questions deserve a reply, who handles sponsorship and press, what you treat as spam, and any subject you never discuss in the comments. Draft replies copy the tone you describe here.

  4. Choose which agent runs it

    NoClick's built-in models work out of the box — or bring Claude Code, Codex, and other coding agents on your own subscription.

  5. Watch it handle a test run

    A staged conversation against a simulated world — then it’s live.

About this agent

A channel past a few thousand subscribers stops having a comments section and starts having an inbox nobody opens. This agent reads everything that came in overnight, counts the repeated questions, digs out the sponsorship and press comments buried under strings of emoji, and reports spam as a wave instead of one comment at a time. Every reply it writes is a draft sitting in Slack, which the creator posts from their own account, or does not.

What people use it for

  • The same question, counted - Eleven people asking which bit you used is not eleven replies, it is a pinned comment or thirty seconds in the next video. Counting is what turns a comment section into a content brief.
  • Sponsorship that stops getting buried - A brand asking about rates leaves a comment like anyone else and sinks under four hundred others by lunchtime. It arrives in its own section with the account and the exact ask.
  • Spam waves handled as waves - Impersonation scams arrive in the hundreds and target your viewers, not you. Seeing the pattern, the count and the videos hit means one clean moderation pass instead of a week of whack a mole.
  • Old uploads that come back - A two year old tutorial getting recommended again brings a fresh audience to advice that has since expired. Knowing that happened overnight is how you catch it before the corrections turn into arguments.

Before you fork

What does it need connected?

The Google account that owns the channel, connected to YouTube, plus the Slack channel where the morning list should land. It uses read operations only: your channel, your videos, and the comments on them. Write the channel rules in your own voice and Test Run will produce a full morning list against a staged batch before you schedule it.

Why does it draft instead of replying itself?

A reply from a channel is read as the creator speaking, and audiences spot an automated one quickly, so drafting keeps the voice yours at the point it counts. No replying, moderating or deleting operation exists in this workflow, only reads and one Slack message, which also means a bad judgment call cannot become a deleted comment from a real fan.

We get thousands of comments a day. Does that break it?

Cost is one run each morning and rises with how many comments have to be read, so a very large channel is a real but predictable expense. Keep it sane by scoping the rules to recent uploads and the videos you actually want watched, rather than the entire back catalogue. Channels with heavy volume often run it twice a day instead, so neither pass has to swallow everything.

Run it with your coding agent

Works with

More agents like this

Browse all agent templates →

Put YouTube Comments Agent to work

Free to start. Guided setup, a test run against staged conversations, and it's live.