Hermes Jarvis + Obsidian Memory is INSANE!
AI News Today | Julian Goldie Podcast
October 15, 2026
Hermes Jarvis + Obsidian Memory is INSANE!
Speakers Julian Goldie
Julian Goldie (0:00)
Today, we're going to be answering some of the latest questions on our Gentic operating system that can basically build and automate anything. And this is using a combination of Hermes and Claude agent and all our agents plugged in together, including even Muse Wonder. And so what we've built here is like an amazing system where you can basically plug in all your agents, you can automate anything. I mean, you can see some of the examples right here even shows you how you're performing across marketing and all your best systems. We can use Hermes agent. We've got, for example, a voice agent. We have even like pop-up agents that are ready to go at any time. So you can see on the right hand side here, we've got a content analysis agent, one that actually keeps up the latest AI automation news. And this is all plugged into our memory systems. So when a new model, for example, like Muse Wonder drops, no problem. We just plug it into our memory system here, and it knows everything about me and every project that we're working on from day one using the second brain system. And so we actually share this whole set up inside the AI Profile Boardroom. And every day I get new questions about how to use the agent, how it works, etc. And that's what we're going to run through today to help our community members as much as possible, and also to help you because if you're watching, you probably have similar sort of questions to what they're asking. So let's just get straight into it and start answering them. So the first question we got here is from Marco. Shout out to Marco, absolute legend in the AI Profile Boardroom. And Marco says, how do you make your Obsidian Vault available to a server based AI agent? So Marco says, I use Obsidian on my Mac with Obsidian Sync. My AI agent, Hermes and Jarvis, runs on a Linux server and should use automated workflows even when my Mac is off. What's the best way to keep selected vault folders on the server and sync the agent's results back without conflicts or data loss, right? Is Obsidian Headless Sync the right approach? How do you handle right permissions, backups, and simultaneous edits by the user and agent? So when it comes to using Obsidian, and I sync it across a couple of different devices, you can actually just use the sync and MCP.
So they have an MCP that allows you to sync across multiple different devices, or you can have the whole vault or individual folders. So I know you're asking, do you want multiple different folders given access to your agent, or for example, the whole vault? And I think this is probably the best way to do it. And then for me personally, I don't really worry too much about what permissions my agent has, because it's quite easy to restore it on the original agent, and we can get that back. And also it's never really been an issue. But yeah, I think the sync and MCP system work really well for you. And this can work across many different devices. You can add a device by scanning the QR code, and then each device has its own token, and it can be disconnected separately, which is pretty cool. And this works with any AI agent, so whether you're using Hermes or any other agent, this would work pretty well. So I think that's probably one of the best ways to do it. Marco says, when of the day Jarvis is live as well. So it sounds like Marco is having a productive day.
Congrats Marco, shout out to you. That's absolutely awesome. This is really good to see people getting results for this, and just seeing people implementing themselves. You're wondering what this is. Basically, it's a voice agent that syncs with Hermes, as you can see right here. And then you can plug in your existing codec subscription to this, so you don't need to pay for the API for it. And it can also use computer use. You can use it whilst you're using any other part of the agent OS. And so, for example, we go to Hermes here, we've got the chat, but then we've also got the voice agent. If we go to Mission Control, we've got the voice agent still over there, and we can use it anytime and speak to it. And it can also, for example, open up apps or control our computer and whatever, right? Really powerful stuff.
And that's it. That's built inside the agentic OS system. So we've got another question here from John. Shout out to John. And John says, I'm an entrepreneur and AI strategist. What I want from this group is to compare notes for people delivering AI for clients. How do you scope it, price it, package it, and what's actually worked? What I can actually do is share with you our SOP and what's worked for us. And we also have a full step-by-step course inside the classroom for agents plus growing an agency.
7 more minutes of transcript below
Thousands of transcripts fetched by people building searchable podcast archives
Fetch the whole transcript
The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000796442971