The MYOS system

Make AI work like it actually knows you.

Five plain folders that turn any AI from a clever stranger into a team that gets you, your work, and how you like things done.

No code No terminal Any AI
Why your AI feels generic

Your AI is brilliant. It also has amnesia.

Open a new chat and you re-explain who you are, what you're working on, how you like things. So it answers like it's meeting you for the first time — because it is. You're not getting help. You're onboarding a new hire, over and over.

The fix is almost too simple
It can't remember. But it can read.

Put your context into a few plain files and point your AI at them before each task. Now it shows up already knowing you — your voice, your projects, your standards. The files are the memory it never had.

Meet MYOS — five folders

Five folders. One simple question each.

Keep them on your computer. Point your AI at them. That's the system.

1

Core

Who you are and how you work — your identity, your voice, your rules. Start here.
core/about-you.md
2

References & Playbooks

What you know and how you do it — your knowledge plus reusable step-by-step playbooks.
playbooks/launch.md
3

Output

Your projects and the work they produce — one folder each.
output/your-project/
4

Notes

What the system learns over time — what worked, what didn't, project by project.
notes/lessons.md
5

Agents

Your AI team — named roles, each with a job. A writer, a researcher, a strategist.
agents/writer.md
How MYOS works

Read before. Save after. That's the loop.

Every chat your AI forgets you. The five folders are its memory: it reads them before a task, and saves a line back after. Each loop, it gets a little sharper.

The MYOS loop in four steps. One: every new chat, the AI starts from zero and forgets you. Two, READ: before a task it reads your five folders — Core, References and Playbooks, Output, Notes, and Agents. Three: it shows up already knowing your voice, your work, and your standards. Four, SAVE: after the task it writes one line back to the Notes folder, so the system gets smarter over time. Then it repeats, a little sharper each loop.

CoreWho am I?
References & PlaybooksWhat do I know?
OutputWhat am I making?
NotesWhat did we learn?
AgentsWho's on my team?
Starts from zero A new chat. It forgets who you are.
READ — load context. It reads your folders, so it shows up knowing you.
SAVE — learn. It writes back what worked, so the system compounds.
What you put in, what you get out

Plain files in. A team out.

You feed it

ContextMemoryConsistencyOwnership
Plain files you own — on your machine, not locked to any single tool.

It gives back

AutonomyLeverageCompounding
A team that runs without you, reuses your playbooks, and gets sharper every week.
The only habit that matters
Point your AI at MYOS before every task.

That's the whole discipline. Do it, and your AI stops being a stranger — for good.