A pad of scrap paper beside the keyboard has done more for my throughput than most of the tooling I changed this year. The minutes a model spends building are the minutes to spend deciding what happens after it lands. - The next sequence - the two or three work orders after this one, written while this one runs. - Notes on the output - what to keep and what to challenge, marked as you read rather than after. - Designs - page layouts, folder trees, flows. Faster to draw than to describe. The dead time in an agentic session is real. A job runs for a few minutes and there are three things you can do with that window: watch it, interrupt it, or think. Watching teaches you nothing and interrupting usually costs a rerun. Writing by hand does something a second chat window does not. The keyboard pulls you back into the work; paper keeps you above it, which is the seat you are supposed to be in while an agent is running. The reading half matters more than the planning half. Marking what to challenge as an output comes in, rather than reaching the end and reconstructing it, is the difference between a next instruction that corrects course and one that just says carry on. The bit I have not solved is getting any of it back into the system. Everything on that pad is orientation the workspace never sees. One of the synced writing tablets pointed at a folder would close that - handwriting arriving as context on the next prompt, and the notes logged as part of the record of how a build actually went. If anyone here has wired one up, I want to hear how. Genuinely think that is one of the more interesting unbuilt pieces sitting in plain sight...

This is an ICM like repo https://github.com/obra/superpowers?utm_source=join.theaigent.xyz&utm_medium=newsletter&utm_campaign=superpowers-turns-claude-into-a-method-not-a-chat&_bhlid=b0c12695cc379e6559f600b15279be31dae6b52c

I just watched one of Jake's breakdowns in Foundations, and the biggest thing that stood out was how much time is wasted between the call and action: manual summaries, rework, lost context. The math for a 5-person team, 3 calls/week: - 30 mins per person per call to summarize/document = 2.5 hours per call wasted - 3 calls/week × 2.5 hours = 7.5 hours/week - 52 weeks/year = 390 hours reclaimed annually The ICM app: Transcription → structured notes → action items → validation → output. One workspace template, runs every call, shared guardrails across all projects. Who else sees this as a good AI app to build for a team setting?

The workflow: Hermes (orchestrator) routes every request. Skills already declare which model tier they need (the model: frontmatter is real, not decorative). When a request touches sensitive data, Hermes defaults to the local model — Gemma, via the Ollama instance Hermes already runs, currently idle — instead of escalating straight to Claude. 1. Local model scopes the request — reads the sensitive material, decides what needs coding. It only proposes; it never makes the final call. 2. A deterministic script does the actual encoding — swaps real names/values for stable per-entity codes via a codebook (table lookup, not AI judgment). . 3. A preflight check verifies nothing raw survived before anything leaves the machine — the check that can demonstrably fail on a bad run, not just a claim that coding happened. 4. Only then does it ask permission to send the coded payload out — your "greenlight and asks permission" step. 5. Claude works entirely on coded tokens — never sees a real name, real entity, real number. 6. The same local codebook decodes Claude's answer back to real names before you see it. That's the full round trip. Claude never sees a real name; you never see a code. What am I not thinking of?

🙋 A little about me: AI Engineer, implementing multi-agent systems. Pivoting to ICM. 🎯 My current goal: To learn as much as possible. Share my own thoughts and read yours. Discussion is healthy for everyone. 💪 What I'm currently building/working on: I'm creating a restaurant order automation. All communication is done through WhatsApp (clients and staff). Clients can schedule orders while talking to the bot, and the bot can message the delivery driver for confirmation, or even the cashier/manager/owner for any context validation. 🤔 My biggest struggle or question right now: I'm using Kafka right now in the current version of the system. The idea is to offload tasks to external processes. However, I'm still seeing if there is a cleanest of solving the problem. I like the auditability of Kafka, just worried it might be overkill. Let's get it ICMers 🚀