The practice.
AI writes a lot of code, very fast, and gets some of it wrong. Software teams have been catching wrong code for decades with tests, specs, shared knowledge, and decent ways of talking to each other. The tools below are how I gave AI the same scaffolding.
If you'd rather see what I've built with this practice first, go here →
-
Doable Doable
My software team is me and a selection of AIs, one Claude per project. Some of those projects are tools the others use, so the bots need a way to talk to each other. Doable is it: a task list per bot, and any bot can raise a task on another bot's queue. I'm on the team too, so I have a queue there as well. When a bot needs something only I can do, root work on the host or anything physical, it asks me.
-
vncPool + adbPool vncPool + adbPool
vncPool gives a Claude a real Linux desktop with a real browser inside it, so it can drive the feature and watch the result. Reading what is on screen is its own problem, since the screenshots Claude gets back are low resolution, so there is an OCR tool and a zoom tool sat on top. adbPool is the Android sibling, a pool of emulators so several bots can test mobile apps at once.
-
Quality Quality
With a bit of coaxing, AI writes good code. Quality is the bit that does the coaxing. It scans each project for the things worth catching early: secrets and credentials sat in the wrong place, basic security issues, god classes and oversized modules, structure that will be a pain to work with later. The findings go back to the bot that wrote the code so the first cleanup pass happens on its own. By the time I'm reading, the obvious stuff has already been dealt with.
-
docVault docVault
docVault is a shared wiki any AI on the server can read and write to. Claude is not the only one running on hp. Codex, OpenCode and a handful of others all reach for the same docVault. The rule book lives there: how to set up a new project, how to register a backup, how to write a test that actually proves something. Situational guidance lives there too, the stuff I don't want in every prompt but do want pulled in on demand. When one of the AIs learns something the others would benefit from, it writes learnings into the docVault.
-
Lore Lore
Lore is the library of every conversation I have with an AI. Claude, Codex, OpenCode, anything, all semantically searchable. When I half-remember a discussion from three weeks ago and can't find which project or session it happened in, I ask Lore. Lore almost always finds it, I drop the session ID into the relevant tool, and I'm straight back where I left off. Bots have an MCP access there too if they need to find out how another bot handled a particular topic.
-
roadTerm roadTerm
roadTerm is my terminal. Each Claude session lives in there, persistent across devices. I start something on the desktop at home, carry on from my phone on the train, and pick it back up on the laptop in the office. Same session, same open files, same Claude waiting where I left it. The other thing I lean on more than I expected is voice. Most of the time I'm not typing at Claude, I'm dictating into roadTerm and letting Whisper turn it into the prompt. Long thoughts come out faster that way, and Claude is good enough at distilling them into a plan/spec I can use.
-
My part in it
In this setup I'm more like a product manager than a coder. With so much code being written I'm not reading every line, but I read the parts that decide whether the system behaves: inputs and outputs, the shape of a Docker container, where secrets sit, the seams between services. Quality covers what I used to scan for by hand, so when I do read code I'm reading for judgement and direction rather than for typos. A fuller write-up of how this side works is on its way.
