Hey guys, Been running the folder structure for a while now — works great. I added one more layer for client projects: sub-agents get spawned inside the client folder, giving you a zero-bleed architecture where they literally can't touch anything outside their scope. A separate agent handles knowledge base access and passes relevant info down to them. Clean separation, no cross-project contamination. This works pretty well. But: Regarding David's 5 Levels of AI Mastery (here in the classrooms) folderstructure + CLI is level 3. Who do we get further? Level 4: Multi-Agent Systems David talks about naming the agents, and only talks to the orchestrator, which hans the tasks over to another agent. My approach: since we're using Skills instead of named agents, I'd just drop something like this into CLAUDE.md: "You are an orchestrator only. Spawn sub-agents for every task, as many as needed. Make sure each one has access to the right skills and context." Is that what David means — or is there a better way? Concrete use case: I've got a 30h course project with outline and skills already built. Do I just tell Claude to slice it up and spawn sub-agents freely, or does it make more sense to hard-code the orchestration. somehow? Level 5: The Factory (the team runs without you) Level 5 is defined as " you explain the goal, go to sleep and wake up a fullfilled work. In my opinion, you could go any further, and let an agent review the status of all projects and start the work on his own (e.g. checkiing Meta ads -> deleting bad ones, creating new ones). Right now, I am running on my macbook. But I think the next step would be to deploy it to the cloud, so n8n workflows and scheduled tasks in my workspace would live in the same place. Where would you host it — VPS, Supabase? Which SDK — Claude Agent SDK, Anthropic SDK, custom dashboard? And how do you handle security at that scale?

So if I understand correctly, the Hermes agent is the Claude agent that I’m talking to in VS Code. So….. for business owners… a self governing business is the holy grail of what we’re doing here, right? A self governing company starts at VS Code and gets built brick by brick, and that’s what’s happening right now? That’s what we’re doing? And in like 6-12 months we’ll have amassed enough data and infrastructure to support that? Am I getting it?