Skip to content

You weren't meant to live at your computer

You talk to Claude.
Your Clone does your work.

Your Clone is your own AI, trained on you and the tools you already use. Say “tag” and it goes to work, calling in a team of expert Agents, AI workers that each own a job. You step away from the screen and get your time back. The work keeps going, and the results arrive when you come back.

See how it works

You stay in control. Nothing foundational changes without your sign-off.

Coming soon for our favorite early people. Get on the list.

Trained on you
EmailCalendarDocumentsMeetingsWeb & SocialYour chats
say “tag”

Your Clone

a second you, that works

A team of Agents goes to work

expert workers your Clone calls in

Your work, done
The bet
Most people think

AI is a smarter tool that helps you get through your work.

The real shift

The real shift is you and your Clone, working with a team of Agents. You build, decide, and refine together, and each cycle is stronger than the last.

Not “AI helps you.” A second you that does the work, so you can step away.

How it works

Say “tag.”
Your Clone takes it from there.

You keep talking to Claude, the same chat you use now. When you want something done, you tag in your Clone. It works in your voice and calls in a team of Agents, the expert AI workers that each own a job, each powered by Skills, the packaged know-how that lets them do it. Even a lightly trained Clone is useful on day one, and it gets sharper every time you use it.

  1. 01

    You tell Claude what you want

    Just say it in plain words, the way you would to a teammate. No prompts to engineer.

  2. 02

    You tag in your Clone

    One word hands the work to your Clone, in your voice, with your rules about what it can and can’t do.

  3. 03

    Your Clone goes to work

    It calls in a team of Agents and gets things done, while you close the laptop and step away.

  4. 04

    It reports back

    The work continues without you. Results arrive when you come back, with the receipts, not just “done.”

Claude chats. Your Clone works.

The one number we lead with

We measure one thing first:
Human Hours Saved.

Not tokens used. Not tasks run. The number that matters is the time the system gives back to you, away from screens, for the work and life only you can do.

And we count it honestly. It includes the time you would otherwise spend working with the AI: the supervising, the prompting, and the waiting too. Most tools quietly leave that out. We put it first.

What we report, in order
#1 · reported first

Human Hours Saved

the time you get back

Tokens used

a gauge, not a gate

secondary

Tokens and compute are a soft awareness for visibility, reported beneath the hours. They never block your progress.

How it knows you

It learns you, and keeps that knowledge honest.

Your Clone’s memory has three parts. Every piece is tagged with where it came from and checked against reality. Nothing important changes without your say-so.

read first

Active

What’s true right now

Your profile, your goals, today’s plan. The current operating picture your Clone reads first, every time.

Kept sharp and current

the capture layer

Adaptive

Where new signals land

New signals from your tools land and get sorted here. It runs daily today, and is built to go continuous when live connectors and scheduling arrive. Reversible for seven daily cycles, about a week.

Tagged with its source

your sign-off
for keeps

Archival

What’s confirmed for good

Durable, confirmed truth. Edited rarely, swept for staleness, and never changed without your confirmation.

Checked against reality daily

Nothing becomes confirmed truth without your sign-off. A signal is captured, checked against what you already believe, and only when you approve does it stick. No silent overwrites, ever.

How it handles everything

Every action, handled the same three ways.

When anything surfaces, a request, a signal, a problem, the system decides what to do with it, independently and in any combination.

Run

Do the work.

Right now, scheduled for later, or queued for the next daily cycle. Always one of those three, never vague.

Record

Write it where it belongs.

Into the right place in memory: activity logs, your profile, your plan. The system never forgets what it did.

Report

Tell you what matters.

Now if it’s urgent, or in the next Daily Digest, your short morning summary of what changed and what needs you.

These are verbs the system does, not categories of objects it stores.

No access is a Report, not a workaround. When the system can’t reach something, it says so plainly instead of improvising around it.

You stay in control

The Review Gate

Synthesis requires a Human. Always.

You stay in control by design. What needs your approval runs on three simple tiers, so you are never buried in rubber-stamps, and never surprised.

Above every tier · never approvable

Some things are off the table for everyone, including you. A final safety scan can block a change that would leak a secret or private data, even after you say yes. Approval is necessary, never enough.

Tier 3Asks you first

Anything foundational, irreversible, or outward-facing: your profile and goals, confirmed truths, anything sent, posted, booked, or deleted outside the system, any new worker or file, and anything touching money or other people’s time.

Tier 2Does it, then tells you

Important background work: real work completed, useful findings, drafts ready for your review. Done on its own, then reported in your next Daily Digest.

Tier 1Just happens, quietly

Routine, reversible housekeeping: working notes, captures, logs, drafts in progress. Reversible for about a week, so a mistake is easy to undo.

When it’s unsure, it goes up one tier

How a change reaches you

When your Clone wants to change what it knows about you, it stages a draft bundle: every proposed change, grouped and explained, waiting for a single yes.

Nothing wrong gets remembered. Nothing silent overwrites you. That is how TagTeam stays trustworthy as it scales.

It gets better, with you

Every cycle, it proposes one way to get better.

The system doesn’t just run. At the close of each cycle it runs Self-Learn, the four S’s, and brings you at least one improvement. You approve. It evolves. Never on its own.

01

Signal

What tripped? What recurred?

Catch the moments the system stumbled or a pattern repeated. The raw material for getting better.

02

Sorting

Was the call right?

Re-examine each Run, Record, or Report decision. Right action, right timing, right place?

03

Scoring

What mattered most?

Ranked by externally-verifiable criteria, not the system grading its own homework. Real outcomes only.

04

Sending

Propose the fix.

Send the improvement to the exact place it belongs, as a draft, for your yes. Never applied on its own.

It all happens in plain language and memory you can read, never in hidden model weights. Every change is visible, and reversible for about a week.

Real and candid

Here’s what’s live today, honestly.

We hold ourselves to the same candor the system uses with you. So here’s the straight version, with no overclaiming.

Live today

The cycle runs daily, triggered. Each day your Clone captures what changed, makes sense of it, checks itself against reality, and sends you a Daily Digest.

Built and ready

Continuous, always-watching operation switches on when two things land: live connectors to your tools, and round-the-clock scheduling. The engine is already built for it.

Until then, anything we run to fill the gap is a named, temporary scheduling bridge: deliberately thin, and cleanly removable the day the platform makes it unnecessary. We name the bridge so no one mistakes it for the destination. We would rather tell you exactly where we are than dress it up.

The discipline

What we refuse.

The discipline is the product. Everything we refuse to build is why the rest works.

No code.

Operators don’t write code. Engineers handle rare edge cases.

No always-on dependence on servers we hand-roll.

Any scaffolding we run ourselves is a named, temporary bridge: deliberately thin, cleanly removable, retired the moment the platform makes it unnecessary. We name the bridge so no one mistakes it for the destination.

No theater.

No decorative complexity, no AI-slop output, no framework accumulation. Every line earns its place.

No drift.

Reuse our vocabulary, concepts, Skills. Foreign frameworks get principles extracted in our terms, never imported wholesale.

No ungated foundational changes.

Anything foundational, irreversible, or outward-facing asks the Human first. The full three-tier law lives at the Review Gate above.

No bare questions from AI.

When an AI surfaces a decision for a human, it brings solutions and options, never just a question.

No data flowing where it shouldn’t.

Security is foundational, not an afterthought.

The promise

You will get your time back. Not by being replaced, by being amplified.

Your Clone does what it’s told, learns from what happens, and surfaces what matters, so you spend your time on the work and life only you can do. Off the screen, where you belong.

Coming soon for our favorite early people. Get on the list.