Hermes Jarvis is mind blowing... artwork

Hermes Jarvis is mind blowing...

AI News Today | Julian Goldie Podcast

June 11, 2026

I Built a Jarvis-Style AI Agent That Can Build Apps, Browse the Web, and Remember EverythingThe script demos “Hermes Jarvis,” a Jarvis-style AI agent the creator is building and improving daily, showing voice interactions like generating a galaxy visual, recalling weekly work, telling an AI joke,...
Speakers: Julian Goldie
**Julian Goldie** (0:00)
This Claude and Hermes setup is going to change the way you use AI forever. Right now, most people are talking to one AI at a time, one tab, one answer, one opinion, and then they're getting pretty mediocre or average results. But what if you could put Claude, Hermes, Gemini, OpenClaw, AntiGravity and more all in one room, all talking to each other, all bouncing off each other, critiquing each other's work and actually taking their ideas and then plugging that into a system that goes off and implements it automatically. That's exactly what I want to show you today. It's called the AI Agent Mastermind. It's the most powerful way to use your AI agents that I've ever built. And instead of one answer, you get six. Instead of generic advice, your agents already know everything about you using the new memory system before you type a word. And instead of just getting ideas, your agents actually just go off and build it for you. So stick with me to the end because I'm going to show you the one thing you absolutely have to set up inside this system. And without it, your AI agents are just going to give you half the results they should be. So let's get into it. I'll show you exactly what I mean. Today I want to show you exactly how you can combine multiple different AI agents together in a kind of group chat system so that you can get the boast of all of your AI agents and orchestrate them in like a simple chat process. I'll also show you a more complicated system that you could use in a second, but this is a powerful way to organize Hermes, to organize Claude, to organize whatever AI agents you want to use together. So let me show you an example of this. I call it the AI Agent Mastermind, which is basically a live group chat with all the agents, each one with a different model. They read your vault, see your Obsidian vault or your memories, then they can reply and turn. They can actually bounce off each other in terms of ideas. They can save notes or start projects and you can just tag on it. So let's say, for example, you want to have Claude and Hermes come up with ideas. So we can say, okay, at Claude, and I come up with some keyword research for SEO about Hermes Agent, right? And so you can see here that it's now gonna read the vault, the Obsidian vault, and then come up with ideas, as you can see here, right? And so it comes up with the keyword research, and these two agents come back with an answer, which is pretty cool. Now, if we said, okay, now come up with content outline for those ideas, e.g.
let's say, for example, we want to go with automation AI, we can continue to chat here. Now you might be wondering, okay, what is the vault? How does that work?
So let me show you an example of this. So if we go to our Obsidian vault, and by the way, you can see our agents are all chiming in now. So we got Gemini, we got Codex, everything else working together. If you go to our memory section here, this is basically pulling in from our memory galaxy and Obsidian itself, so that you can look at, okay, what are we working on recently? And then how can we come up with better ideas and topics around what we've been working on? And so you get very personalized responses inside the chat, and also you can get all of your agents working together. So, for example, here FreeClaude code is responding to OpenClaude, and then we have Codex that's bouncing off Gemini. And then, for example, Hermes gave us some ideas on that SEO content. So Hermes, okay, this is pretty interesting. I want to explain this without confusing anyone watching. So if you look at this, we said Hermes and Claude come up with some keyword research. They both came up with keyword research. Perfect. Now I said come up with a content outline. Claude comes up with a content outline, then Hermes actually critiques it and gives us its own idea. From there, Gemini and Codex actually replied as well, and OpenClaude and FreeClaude code chimed in with their own ideas as well. Now, here's where it gets really interesting. So they save notes to our Obsidian Vault, as you can see here, and then also they will add ideas to our pipeline, which means that if we check out our pipeline over here, it's coming up with new ideas based on those topics.

18 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/1000772152200