Hermes can run a cron job without ICM. It can wake up on schedule, read a source, write a report, send a message, and save the run output. It also has its own durable storage: - session history, - persistent memory, - cron definitions and run logs, - skills, - and whatever files it creates in the workspace. That is useful. But it can still leave you with a familiar mess: > a chat somewhere > a cron log somewhere else > a report with no obvious home > and an agent that needs you to explain the project again next week That is where ICM comes in. I use ICM around my Build Market Close signal cron so the **job is not the whole system**. The cron is just the heartbeat. The structure holds the memory: ```text pinned-post-curriculum-signal/ ├── CONTEXT.md ├── signal-register.md ├── reply-drafts.md ├── curriculum-outline.md └── runs/ └── YYYY-MM-DD/ └── signal-review.md ``` Here is what each part does: - **`CONTEXT.md`** tells a fresh agent what source it may read, what counts as meaningful signal, what it may write, and where it must stop for me. - **`runs/YYYY-MM-DD/`** keeps a dated receipt of every review: what the job saw, what changed, and what it deliberately did not do. - **`signal-register.md`** is the one place for real questions, objections, implementation friction, and learner needs. - **`reply-drafts.md`** holds proposed answers for review. Nothing gets posted automatically. - **`curriculum-outline.md`** turns repeated questions into a better lesson sequence instead of letting them disappear into comments. So the cron does the recurring check. But ICM makes the check **usable**. A fresh agent can open the folder and answer: 1. What is this job for? 2. What source is it allowed to use?

Anybody built an ICM representing the idea of super large initiatives going on in an organization and smaller less impactful initiatives going on in the organization (that move a little bit faster than the large ones) and making sure that the dependencies across all of them are not overlapping significantly? When they do, they get called out so we can find them. Something that highlights and manages, to some degree, the interconnections between different initiatives

Witnessing a friend who's less skilled at technology and AI burn through 5 hours of usage in right about 2 hours created a situation where we had to pause and figure out what was wrong. It did not appear she is doing anything extreme: reading some websites, building some content on what was found, and a few other things. When it gave her the warning, we were a bit surprised. Turns out, a couple of days before, when setting up her account with Claude and installing Claude Desktop, at some point Claude asked, "Do you want to install plugins for sales and plugins for marketing?" and something else. Being a small business owner with little understanding of what that means, she simply checked both boxes for sales and marketing. This triggered Claude desktop to install a series of MCP serversbconnections approximately 25 to 30. Many of which were services you would probably never use. These appeared to be standard Anthropic plugins. Nothing seemed out of the ordinary. Well, except for the usage that just kept climbing to the degree that, before we could fix the problem, she had added credits to the tune of $60 or $70.. Just to get through the tasks that she needed to do . When we realized this, we instructed Claude to explain how to remove these because I had seen in other situations where removing the plugin did not necessarily remove the underlying infrastructure, like an MCP configuration. That said, this time they were lucky, and just removing the plugins did ultimately remove these extra MCPs, but the thing that's important to know Is it that every MCP listing, whether you're authenticated to it or not, costs tokens to maintain in your session? Based on this experience, I'm not suggesting anyone install the straight-up plugins. Tell them to only ever install an MCP or, potentially, a plugin, so that you know exactly what it's installing and whether you're going to need those things or not. Just suggest not realizing the impact. And it's not really in Anthropic's best interest to tell them the situation, because when the result of not knowing triggers the ability to earn double the tokens or get a client to pay more for extra tokens to finish jobs. We calculated that the extra MCPs configured and never used, not even authenticated, were costing approximately 1,500 to 2,500 tokens across the board.

Im considering going live later tonight to record a video talking about edubaware at a 101 level. Would anyone be interested? Update: couldn’t make it tonight but will do one tomorrow night 9/15 at 10:00 PM EST

After 5 months of being in this community, I finally caught the ICM flu... I didn't catch it the first month--we were just having a good time back then. I didn't catch it the second month, either. Those were the "lego" days, and still long before it would really click. I didn't even catch it when I started a second company 6 weeks ago, that I built after being inspired right here in this community. For 5 months, I was vibe coding. Up late, up early, like a middle schooler with a new video game, like the internet just became a thing for the first time... I thought that was the thick of it. No. Oh no.. No, no, no, no, no, no. There are levels, boys, there are levels... I caught the ICM flu once I made a widget that an actual client used. And then it finally clicked. Software. Agents. Human. You can just build a whole foundation for any company using this method. Which, sure, yes, we know that. But, realizing that once you got it... once you've built that first widget, that first workflow tied to a piece of software you coded and an agent you made, you realize... that's the whole thing. Do it once, do it a thousand times. Recycle, reuse. We're all software developers now, baby! About 2 weeks after it finally clicked, I joined the Lyceum because I understood the value of this finally. It does not matter what you plan to do--these skills transfer to ALL SORTS of things. Software, Agents, Human... just apply it to.... anything! As a side note, I have already received value after the first Lyceum session. I finally can see how I can transform my little house cleaning company into a mega machine and create the dream vision I have for this company. FOR ANYONE THAT SAW MY ANNOUNCEMENT ABOUT STARTING A COMPANY... This is going to sound nuts, I know, but Im shutting down the company I just made despite the fact that it's working. I can now see that the greatest play, and most value I can create, is going to come from the company that I have so much experience in already, and not the one that I have limited experience with. My judgements and tastes are strong where I've spent a ton of time already. Really, I have 3 lanes I've spent the majority of my life in, and I've decided I need to see those through at this point because I have way too much specialized knowledge to not do something with it. *And that's the secret sauce.*