Claude + Paperclip Agent Swarm is INSANE! 🤯 artwork

Claude + Paperclip Agent Swarm is INSANE! 🤯

AI News Today | Julian Goldie Podcast

June 11, 2026

How to Orchestrate Teams of AI Agents with Claude (Group Chat + Idea Factory + Agent OS)The video demonstrates how to orchestrate teams of AI agents with Claude so they can brainstorm, critique, and move ideas to implementation with minimal human input.
Speakers: Julian Goldie
**Julian Goldie** (0:00)
Today, I want to show you exactly how to orchestrate your teams of agents we've clawed. And there's two powerful ways that I know to do this directly. And you can see everything that our team has built here. So they've built some awesome stuff. And there's some more examples of this. If we go to the pipeline over here, we can see everything that's been built with our gallery. And there's some pretty amazing stuff. Now, I didn't automate, I mean, I didn't touch any of this. And what I mean by that basically, is when we've created all of this cool stuff, I didn't come up with the idea, I didn't create it, I didn't even tell them to create it. I literally just clicked the button to approve it. And that's the game that we're playing right now, which is basically AI agent teams working without us. So I'll actually show you multiple ways to do this, some of my favorite ways, and then you can organize and decide which one works the best for you. So let me show you the first way, which is I have a group chat, I call it the AI agent mastermind over here. And I can tag different AI agents together. And then from there, they can debate and decide on the next step forward. So let's say for example, we're like, okay, Claude and Hermes, I need to build a powerful app for SEO, come up with some ideas, right? And so what's going to happen now is that we've tagged in Claude, and we've tagged in Hermes into the chat, and they're going to bounce off each other. Now, if I say, okay, give me some more ideas, what you're actually going to see here is because we haven't tagged anyone, the whole team of AI agents are going to pipe in together, right? They're all going to work together. These are AI agents in a group chat, talking among each other and helping each other build and find better stuff to create. So what you'll see here, for example, is that Claude added an idea to the pipeline, and Hermes actually came up with this app. So we said, you know, give me some ideas I need to build for SEO, and then Claude gave us an idea, and then Hermes gave us an idea. Now, what's also happening at the same time, and this is why this part is important, which is our lab, this ideas factory, our pipeline.
Essentially, that's going to take new ideas from this conversation, where we're all brainstorming and coming up with ideas together, and plug it over to our capture list. This is the idea capture list. What this essentially does is allow us to take new ideas and automatically list them. So my AI agents inside this team are working together, they're critiquing each other, they're bouncing ideas off each other. You can see that inside this conversation, no matter what I throw at them. Doesn't matter, right? They'll come up with ideas on the spot. They're trained to do that. They are beasts.
And they're not going to put every single idea on the list. So you can see, for example, Hermes, Gemini and Codex all came up with ideas, and then Codex is like, you know what, that's actually pretty good. Let's plug that in. So for example, here we've got an AI-driven content trend analyzer that we could start creating. Also have a big sample of a predictive ranking probability tracker, which could be pretty cool as well. And so those ideas have now gone over to our idea factory from the pipeline. So you see how we're orchestrating our agents. We're also making sure they do all the hard work instead of me doing it. So they come up with the ideas, they plan it out, they find new stuff, and then they plug it into an ideas list, a to-do list, and all I need to do is click a button to get it started, to get the party started. So let me show you an example. So over here, we have, for example, all these lists of ideas. Let me just refresh this, make sure we've got the latest version. And you see all these different ideas over here, Rem. These are ideas from the group chat. And so, for example, they've recommended coming up with an SEO AI Content Intelligence app with predictive keyword ranking. Now, if we're like, do you know what? That one sounds really cool. Let's get that party started. What it's going to do now is start shaping it. And so it's taking the idea and then it's creating a plan to build it out. And so what that means essentially is we're going from idea to implementation without me having to do anything. I'm not thinking about it. I'm not even wondering what to create. I'm not coming up with the ideas. It's being planned out. Now, once it's been planned out, as you can see, it moves from the capture page, from the capture section here, and it moves over to the human gate.

7 more minutes of transcript below

Feed this to your agent

Try it now — copy, paste, done:

curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Works with Claude, ChatGPT, Cursor, and any agent that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire.

Using your own key:

curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000772145291