Intros for The Connection Hub - The Vault 👤 Who I am: (name + where you're based) 🛠️ What I actually do: (the specific work — not "I'm in real estate" but "I run a 3-agent team doing residential resale in Austin") 🤖 What I'm building with AI right now: (your current project, workflow, or the thing you're stuck on) 🎯 What I'm looking for connection-wise: (pick one or two) 💡 Someone who's solved [X] 🤝 A collaborator / accountability partner 👀 Just here to learn from people in my field 🧰 Trading workflows & systems 📬 Best way to reach me: (DM here / comment / link)

I've got a technical issue that needs some outside help. My background is not tech so it's not something I'm confident giving Claude an answer to so I got it to ask the question and said I'd give it any responses I got. Here's what it asked... I help run a small street food business, and we've built our internal tools as an ICM workspace. It's been working well. I've hit a structural question I'd like your view on before we change anything. The setup One folder holds four tools side by side, each a full pipeline with its own CLAUDE.md: - PRODUCTION-RUN: weekly orders in; production sheet and courier/invoice files out - SALES-HISTORY: who's buying more or less, week to week - COST-CALCULATOR: labour and ingredient cost per product - STAFF-HOURS: hours, holiday, and the monthly payroll file to the accountant The top-level CLAUDE.md only routes to the right tool. There's no shared reference folder at the top level. Our rule versus the skill Our top-level CLAUDE.md says each tool "holds its own data, and points at nothing outside itself", and that "a pointer across folders is how a suite turns into a tangle." The icm-architect skill's Umbrella form is looser. Sub-pipelines "don't share state" except through the root reference layers, and shared reference material should be linked up rather than duplicated. What actually happens Two of our tools break our own rule on purpose: - SALES-HISTORY reads PRODUCTION-RUN's weekly production sheets and its customer list. It also keeps its own snapshot of customer prices, and its own copies of older hand-kept sheets. - COST-CALCULATOR reads the same weekly production sheets. In both, only the first stage reaches across, and only once. Every later stage is told not to load anything from PRODUCTION-RUN. STAFF-HOURS reads nothing outside itself. It only mentions PRODUCTION-RUN in its notes, as a comparison. Each tool runs a check at the start of every session that reports any file path it can't find inside its own folder. So every reference to another tool gets reported as broken, every session.

Hey team are there any discussions in here talking about encrypting your files and folders that have been delivered to clients so your work is protected?

hello, has anyone find how to slowly migrate certain parts of their ICM to use jev? things like routing, tool calls and so on. seems like it will be a mish mash in sequence of jev and llm calls to accomplish task in the ICM system which could speed up retrievals? but it will be trouble with llm caching and debugging when things go wrong. what do you think?

A founder asked me a question this week that I couldn't answer on the spot: do other businesses have a system like this? So I went and looked. Short version: outside this room, I can't find them. Everyone here runs some version of it already, a folder of markdown that Claude reads and maintains, and I'd stopped noticing how unusual that is. I'd like the room to tell me where the others are hiding, or whether we're further ahead than we think. The system in question is the one I've been installing at a three-person shop since August. The part that's now real: a knowledge base in a git repo that their own Claude reads before it does any work, a file register that watches their Drive and logs every arrival and every move, and a pitch engine that runs a brief through four human approval gates with the gaps written on the slide. Every filing carries who approved it or which rule filed it. Every page carries where it came from. When something ends up in the wrong place, the record says whether a person put it there or our rule did. None of the pieces are exotic. Markdown, git, a spreadsheet, an Apps Script, skills. What's different is that the knowledge is compiled and maintained, not retrieved. The agent doesn't search a pile of PDFs every time. It reads pages it helped write, with the provenance on them, and the rules for what it may and may not do live in the same repo. Here's what I found when I went looking for who else does this. The big holding companies in that industry have "operating systems" now. Real ones, sold to clients as the reason to hire the group. That's a data platform with agents on it, top down, owned by the group. Not a thing three people could own. Everyone in the middle has AI on top of files. The numbers from this year's surveys: about two thirds of agencies run brief generation in production, a third draft client reports with it, and the single biggest problem the independents report is getting AI into their actual workflows. Every "AI knowledge base" guide I read describes the same thing. Search over the shared drive. A memory server. Chat with your documents. Useful. Also not a record of anything.