Case study
Internal toolUBA Panel
A read-only status dashboard that shows whether the UBA-BRAIN system — clients, skills, health checks — is in good shape, from any device on the network.
FIG. 001 / ARCHITECTURE
How the system connects
Live model
Problem
With a shared knowledge base and skill library synced across three machines, there was no quick way to see, at a glance, whether everything was consistent — which skills were active, which clients were configured, or whether anything had drifted — without opening files by hand on each machine.
System
A Next.js/TypeScript app reads the UBA-BRAIN folder directly and renders its state as a dashboard, running persistently on a home Mac via launchd and reachable only over a private Tailscale network, with a small deploy script to push updates from a development machine to the always-on host.
FIG. 002 / HOW IT WORKS
How it works
A small internal tool built once the main automation "brain" grew complex enough to need its own status view: a read-only panel that surfaces the health of the client/skill knowledge base it's watching, always running on a home server and accessible only over a private network — infrastructure work in service of the larger automation system rather than a product on its own.
Stack
What it runs on
Related systems
Brand Knowledge Hub & AI Listing Generator
Internal toolA local web app holding each brand's products, branding and images in one place, with an AI chat grounded on that brand to generate on-brief images.
TruckPulse
Internal toolAutomated load-profitability and dispatcher-performance monitoring for a trucking company, built from Gmail broker PDFs and Telegram driver-rate messages.
IshFlow
Internal toolInstagram DM automation and lightweight CRM that lets a small seller manage customers, catalog and orders entirely from a Telegram bot.
Want a system like this for your team?
We start by mapping one workflow on a 30-minute call.