This part continues the Handmade ChatGPT you built before — its code is
carried into this folder. Setup only; the build is the next task.

Confirm AGENTS.md still declares, each at the start of its own line:

stack: <language, framework, storage>
run:
test:

They should already be there from models you carried in. Make sure
run: still starts the product and test: still runs a suite that passes —
a part carried honestly starts from working code.

Then plan THIS part in TODO.md: add the steps for accounts, so more than one person can use it, each isolated, in
product terms, and check them off as they land. AGENTS.md and TODO.md are
read by every judge of this part as evidence of how you work.

So far the product belongs to whoever opens the page. Now it belongs to
an account: people sign up, sign in, and see only their own providers,
conversations and settings. You decide when it is done; a judge panel
scores the result.

Scenarios:

Scenario: Signing up and in
When Grace signs up with an email and a password
Then she is signed in, and can sign out and back in with the same password
And a wrong password, or an email already taken, is refused with a reason

Scenario: Yours stays yours
When Grace signs in
Then she sees no provider, no model and no conversation of Ada's
And opening Ada's conversation by its address is refused, not shown

Scenario: Hiding the buttons is not enough
When Grace sends a message into Ada's conversation without using the page
Then the product refuses her
And Ada's conversation is unchanged

Scenario: Secrets stay secret
Then passwords are not stored as typed
And provider keys are never sent back to any page

Everything beyond the scenarios is yours: how signing in looks, what a
signed-out visitor sees, what happens to conversations made before there
were accounts.

When you are done, write .ololo/chatgpt-accounts-done.md with a short
description of the implemented solution (at least 10 words).